feat: upgrade xinference to 0.2.1 which support stream response

Co-authored-by: jyong <jyong@dify.ai>

README.md

@@ -30,7 +30,7 @@ Visual data analysis, log review, and annotation for applications
 - [x] **Spark**
 - [x] **Wenxin**
 - [x] **Tongyi**
-- [x] **ChatGLM**
+
 
 We provide the following free resources for registered Dify cloud users (sign up at [dify.ai](https://dify.ai)):
 * 600,000 free Claude model tokens to build Claude-powered apps

api/Dockerfile

@@ -16,7 +16,7 @@ EXPOSE 5001
 WORKDIR /app/api
 
 RUN apt-get update && \
-    apt-get install -y bash curl wget vim gcc g++ python3-dev libc-dev libffi-dev
+    apt-get install -y bash curl wget vim gcc g++ python3-dev libc-dev libffi-dev nodejs
 
 COPY requirements.txt /app/api/requirements.txt
 

api/commands.py

@@ -20,7 +20,7 @@ from models.model import Account
 import secrets
 import base64
 
-from models.provider import Provider, ProviderType, ProviderQuotaType
+from models.provider import Provider, ProviderType, ProviderQuotaType, ProviderModel
 
 
 @click.command('reset-password', help='Reset the account password.')
@@ -102,6 +102,7 @@ def reset_encrypt_key_pair():
     tenant.encrypt_public_key = generate_key_pair(tenant.id)
 
     db.session.query(Provider).filter(Provider.provider_type == 'custom').delete()
+    db.session.query(ProviderModel).delete()
     db.session.commit()
 
     click.echo(click.style('Congratulations! '

api/config.py

@@ -100,7 +100,7 @@ class Config:
         self.CONSOLE_URL = get_env('CONSOLE_URL')
         self.API_URL = get_env('API_URL')
         self.APP_URL = get_env('APP_URL')
-        self.CURRENT_VERSION = "0.3.14"
+        self.CURRENT_VERSION = "0.3.15"
         self.COMMIT_SHA = get_env('COMMIT_SHA')
         self.EDITION = "SELF_HOSTED"
         self.DEPLOY_ENV = get_env('DEPLOY_ENV')

api/controllers/console/apikey.py

@@ -1,4 +1,5 @@
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 import flask_restful
 from flask_restful import Resource, fields, marshal_with
 from werkzeug.exceptions import Forbidden

api/controllers/console/app/app.py

@@ -3,7 +3,9 @@ import json
 import logging
 from datetime import datetime
 
-from flask_login import login_required, current_user
+import flask
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, reqparse, fields, marshal_with, abort, inputs
 from werkzeug.exceptions import Forbidden
 
@@ -316,7 +318,7 @@ class AppApi(Resource):
 
         if current_user.current_tenant.current_role not in ['admin', 'owner']:
             raise Forbidden()
-        
+
         app = _get_app(app_id, current_user.current_tenant_id)
 
         db.session.delete(app)

api/controllers/console/app/audio.py

@@ -2,7 +2,7 @@
 import logging
 
 from flask import request
-from flask_login import login_required
+from core.login.login import login_required
 from werkzeug.exceptions import InternalServerError, NotFound
 
 import services

api/controllers/console/app/completion.py

@@ -5,7 +5,7 @@ from typing import Generator, Union
 
 import flask_login
 from flask import Response, stream_with_context
-from flask_login import login_required
+from core.login.login import login_required
 from werkzeug.exceptions import InternalServerError, NotFound
 
 import services

api/controllers/console/app/conversation.py

@@ -1,7 +1,8 @@
 from datetime import datetime
 
 import pytz
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, reqparse, fields, marshal_with
 from flask_restful.inputs import int_range
 from sqlalchemy import or_, func

api/controllers/console/app/generator.py

@@ -1,4 +1,5 @@
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, reqparse
 
 from controllers.console import api

api/controllers/console/app/message.py

@@ -3,7 +3,7 @@ import logging
 from typing import Union, Generator
 
 from flask import Response, stream_with_context
-from flask_login import current_user, login_required
+from flask_login import current_user
 from flask_restful import Resource, reqparse, marshal_with, fields
 from flask_restful.inputs import int_range
 from werkzeug.exceptions import InternalServerError, NotFound
@@ -16,6 +16,7 @@ from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required
 from core.model_providers.error import LLMRateLimitError, LLMBadRequestError, LLMAuthorizationError, LLMAPIConnectionError, \
     ProviderTokenNotInitError, LLMAPIUnavailableError, QuotaExceededError, ModelCurrentlyNotSupportError
+from core.login.login import login_required
 from libs.helper import uuid_value, TimestampField
 from libs.infinite_scroll_pagination import InfiniteScrollPagination
 from extensions.ext_database import db

api/controllers/console/app/model_config.py

@@ -3,12 +3,13 @@ import json
 
 from flask import request
 from flask_restful import Resource
-from flask_login import login_required, current_user
+from flask_login import current_user
 
 from controllers.console import api
 from controllers.console.app import _get_app
 from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required
+from core.login.login import login_required
 from events.app_event import app_model_config_was_updated
 from extensions.ext_database import db
 from models.model import AppModelConfig

api/controllers/console/app/site.py

@@ -1,5 +1,6 @@
 # -*- coding:utf-8 -*-
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, reqparse, fields, marshal_with
 from werkzeug.exceptions import NotFound, Forbidden
 

api/controllers/console/app/statistic.py

@@ -4,7 +4,8 @@ from datetime import datetime
 
 import pytz
 from flask import jsonify
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, reqparse
 
 from controllers.console import api

api/controllers/console/auth/data_source_oauth.py

@@ -5,9 +5,12 @@ from typing import Optional
 import flask_login
 import requests
 from flask import request, redirect, current_app, session
-from flask_login import current_user, login_required
+from flask_login import current_user
+
 from flask_restful import Resource
 from werkzeug.exceptions import Forbidden
+
+from core.login.login import login_required
 from libs.oauth_data_source import NotionOAuth
 from controllers.console import api
 from ..setup import setup_required

api/controllers/console/datasets/data_source.py

@@ -3,7 +3,8 @@ import json
 
 from cachetools import TTLCache
 from flask import request, current_app
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, marshal_with, fields, reqparse, marshal
 from werkzeug.exceptions import NotFound
 

api/controllers/console/datasets/datasets.py

@@ -1,6 +1,7 @@
 # -*- coding:utf-8 -*-
 from flask import request
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, reqparse, fields, marshal, marshal_with
 from werkzeug.exceptions import NotFound, Forbidden
 import services

api/controllers/console/datasets/datasets_document.py

@@ -4,7 +4,8 @@ from datetime import datetime
 from typing import List
 
 from flask import request
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, fields, marshal, marshal_with, reqparse
 from sqlalchemy import desc, asc
 from werkzeug.exceptions import NotFound, Forbidden
@@ -764,11 +765,13 @@ class DocumentMetadataApi(DocumentResource):
         metadata_schema = DocumentService.DOCUMENT_METADATA_SCHEMA[doc_type]
 
         document.doc_metadata = {}
-
-        for key, value_type in metadata_schema.items():
-            value = doc_metadata.get(key)
-            if value is not None and isinstance(value, value_type):
-                document.doc_metadata[key] = value
+        if doc_type == 'others':
+            document.doc_metadata = doc_metadata
+        else:
+            for key, value_type in metadata_schema.items():
+                value = doc_metadata.get(key)
+                if value is not None and isinstance(value, value_type):
+                    document.doc_metadata[key] = value
 
         document.doc_type = doc_type
         document.updated_at = datetime.utcnow()

api/controllers/console/datasets/datasets_segments.py

@@ -1,9 +1,8 @@
 # -*- coding:utf-8 -*-
 import uuid
 from datetime import datetime
-
 from flask import request
-from flask_login import login_required, current_user
+from flask_login import current_user
 from flask_restful import Resource, reqparse, fields, marshal
 from werkzeug.exceptions import NotFound, Forbidden
 
@@ -15,6 +14,7 @@ from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required
 from core.model_providers.error import LLMBadRequestError, ProviderTokenNotInitError
 from core.model_providers.model_factory import ModelFactory
+from core.login.login import login_required
 from extensions.ext_database import db
 from extensions.ext_redis import redis_client
 from models.dataset import DocumentSegment

api/controllers/console/datasets/file.py

@@ -8,7 +8,8 @@ from pathlib import Path
 
 from cachetools import TTLCache
 from flask import request, current_app
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, marshal_with, fields
 from werkzeug.exceptions import NotFound
 

api/controllers/console/datasets/hit_testing.py

@@ -1,6 +1,7 @@
 import logging
 
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, reqparse, marshal, fields
 from werkzeug.exceptions import InternalServerError, NotFound, Forbidden
 

api/controllers/console/explore/installed_app.py

@@ -1,7 +1,8 @@
 # -*- coding:utf-8 -*-
 from datetime import datetime
 
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, reqparse, fields, marshal_with, inputs
 from sqlalchemy import and_
 from werkzeug.exceptions import NotFound, Forbidden, BadRequest

api/controllers/console/explore/recommended_app.py

@@ -1,5 +1,6 @@
 # -*- coding:utf-8 -*-
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, fields, marshal_with
 from sqlalchemy import and_
 

api/controllers/console/explore/wraps.py

@@ -1,4 +1,5 @@
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource
 from functools import wraps
 

api/controllers/console/universal_chat/wraps.py

@@ -1,7 +1,8 @@
 import json
 from functools import wraps
 
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource
 from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required

api/controllers/console/workspace/account.py

@@ -3,7 +3,8 @@ from datetime import datetime
 
 import pytz
 from flask import current_app, request
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, reqparse, fields, marshal_with
 
 from services.errors.account import CurrentPasswordIncorrectError as ServiceCurrentPasswordIncorrectError

api/controllers/console/workspace/members.py

@@ -1,6 +1,7 @@
 # -*- coding:utf-8 -*-
 from flask import current_app
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, reqparse, marshal_with, abort, fields, marshal
 
 import services

api/controllers/console/workspace/model_providers.py

@@ -1,4 +1,5 @@
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, reqparse
 from werkzeug.exceptions import Forbidden
 

api/controllers/console/workspace/models.py

@@ -1,4 +1,5 @@
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, reqparse
 
 from controllers.console import api

api/controllers/console/workspace/providers.py

@@ -1,5 +1,6 @@
 # -*- coding:utf-8 -*-
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, reqparse
 from werkzeug.exceptions import Forbidden
 

api/controllers/console/workspace/tool_providers.py

@@ -1,6 +1,7 @@
 import json
 
-from flask_login import login_required, current_user
+from flask_login import current_user
+from core.login.login import login_required
 from flask_restful import Resource, abort, reqparse
 from werkzeug.exceptions import Forbidden
 

api/controllers/console/workspace/workspace.py

@@ -2,10 +2,13 @@
 import logging
 
 from flask import request
-from flask_login import login_required, current_user
-from flask_restful import Resource, fields, marshal_with, reqparse, marshal
+from flask_login import current_user
+from core.login.login import login_required
+from flask_restful import Resource, fields, marshal_with, reqparse, marshal, inputs
+from flask_restful.inputs import int_range
 
 from controllers.console import api
+from controllers.console.admin import admin_required
 from controllers.console.setup import setup_required
 from controllers.console.error import AccountNotLinkTenantError
 from controllers.console.wraps import account_initialization_required
@@ -43,6 +46,13 @@ tenants_fields = {
     'current': fields.Boolean
 }
 
+workspace_fields = {
+    'id': fields.String,
+    'name': fields.String,
+    'status': fields.String,
+    'created_at': TimestampField
+}
+
 
 class TenantListApi(Resource):
     @setup_required
@@ -57,6 +67,38 @@ class TenantListApi(Resource):
         return {'workspaces': marshal(tenants, tenants_fields)}, 200
 
 
+class WorkspaceListApi(Resource):
+    @setup_required
+    @admin_required
+    def get(self):
+        parser = reqparse.RequestParser()
+        parser.add_argument('page', type=inputs.int_range(1, 99999), required=False, default=1, location='args')
+        parser.add_argument('limit', type=inputs.int_range(1, 100), required=False, default=20, location='args')
+        args = parser.parse_args()
+
+        tenants = db.session.query(Tenant).order_by(Tenant.created_at.desc())\
+            .paginate(page=args['page'], per_page=args['limit'])
+
+        has_more = False
+        if len(tenants.items) == args['limit']:
+            current_page_first_tenant = tenants[-1]
+            rest_count = db.session.query(Tenant).filter(
+                Tenant.created_at < current_page_first_tenant.created_at,
+                Tenant.id != current_page_first_tenant.id
+            ).count()
+
+            if rest_count > 0:
+                has_more = True
+        total = db.session.query(Tenant).count()
+        return {
+            'data': marshal(tenants.items, workspace_fields),
+            'has_more': has_more,
+            'limit': args['limit'],
+            'page': args['page'],
+            'total': total
+                }, 200
+
+
 class TenantApi(Resource):
     @setup_required
     @login_required
@@ -92,6 +134,7 @@ class SwitchWorkspaceApi(Resource):
 
 
 api.add_resource(TenantListApi, '/workspaces')  # GET for getting all tenants
+api.add_resource(WorkspaceListApi, '/all-workspaces')  # GET for getting all tenants
 api.add_resource(TenantApi, '/workspaces/current', endpoint='workspaces_current')  # GET for getting current tenant info
 api.add_resource(TenantApi, '/info', endpoint='info')  # Deprecated
 api.add_resource(SwitchWorkspaceApi, '/workspaces/switch')  # POST for switching tenant

api/core/docstore/dataset_docstore.py

@@ -103,6 +103,7 @@ class DatesetDocumentStore:
                     content=doc.page_content,
                     word_count=len(doc.page_content),
                     tokens=tokens,
+                    enabled=False,
                     created_by=self._user_id,
                 )
                 if 'answer' in doc.metadata and doc.metadata['answer']:

api/core/indexing_runner.py

@@ -691,6 +691,7 @@ class IndexingRunner:
                 DocumentSegment.status == "indexing"
             ).update({
                 DocumentSegment.status: "completed",
+                DocumentSegment.enabled: True,
                 DocumentSegment.completed_at: datetime.datetime.utcnow()
             })
 

api/core/login/login.py

@@ -0,0 +1,108 @@
+import os
+from functools import wraps
+
+import flask_login
+from flask import current_app
+from flask import g
+from flask import has_request_context
+from flask import request
+from flask_login import user_logged_in
+from flask_login.config import EXEMPT_METHODS
+from werkzeug.exceptions import Unauthorized
+from werkzeug.local import LocalProxy
+
+from extensions.ext_database import db
+from models.account import Account, Tenant, TenantAccountJoin
+
+#: A proxy for the current user. If no user is logged in, this will be an
+#: anonymous user
+current_user = LocalProxy(lambda: _get_user())
+
+
+def login_required(func):
+    """
+    If you decorate a view with this, it will ensure that the current user is
+    logged in and authenticated before calling the actual view. (If they are
+    not, it calls the :attr:`LoginManager.unauthorized` callback.) For
+    example::
+
+        @app.route('/post')
+        @login_required
+        def post():
+            pass
+
+    If there are only certain times you need to require that your user is
+    logged in, you can do so with::
+
+        if not current_user.is_authenticated:
+            return current_app.login_manager.unauthorized()
+
+    ...which is essentially the code that this function adds to your views.
+
+    It can be convenient to globally turn off authentication when unit testing.
+    To enable this, if the application configuration variable `LOGIN_DISABLED`
+    is set to `True`, this decorator will be ignored.
+
+    .. Note ::
+
+        Per `W3 guidelines for CORS preflight requests
+        <http://www.w3.org/TR/cors/#cross-origin-request-with-preflight-0>`_,
+        HTTP ``OPTIONS`` requests are exempt from login checks.
+
+    :param func: The view function to decorate.
+    :type func: function
+    """
+
+    @wraps(func)
+    def decorated_view(*args, **kwargs):
+        auth_header = request.headers.get('Authorization')
+        admin_api_key_enable = os.getenv('ADMIN_API_KEY_ENABLE', default='False')
+        if admin_api_key_enable:
+            if auth_header:
+                if ' ' not in auth_header:
+                    raise Unauthorized('Invalid Authorization header format. Expected \'Bearer <api-key>\' format.')
+                auth_scheme, auth_token = auth_header.split(None, 1)
+                auth_scheme = auth_scheme.lower()
+                if auth_scheme != 'bearer':
+                    raise Unauthorized('Invalid Authorization header format. Expected \'Bearer <api-key>\' format.')
+                admin_api_key = os.getenv('ADMIN_API_KEY')
+
+                if admin_api_key:
+                    if os.getenv('ADMIN_API_KEY') == auth_token:
+                        workspace_id = request.headers.get('X-WORKSPACE-ID')
+                        if workspace_id:
+                            tenant_account_join = db.session.query(Tenant, TenantAccountJoin) \
+                                .filter(Tenant.id == workspace_id) \
+                                .filter(TenantAccountJoin.tenant_id == Tenant.id) \
+                                .filter(TenantAccountJoin.role == 'owner') \
+                                .one_or_none()
+                            if tenant_account_join:
+                                tenant, ta = tenant_account_join
+                                account = Account.query.filter_by(id=ta.account_id).first()
+                                # Login admin
+                                if account:
+                                    account.current_tenant = tenant
+                                    current_app.login_manager._update_request_context_with_user(account)
+                                    user_logged_in.send(current_app._get_current_object(), user=_get_user())
+        if request.method in EXEMPT_METHODS or current_app.config.get("LOGIN_DISABLED"):
+            pass
+        elif not current_user.is_authenticated:
+            return current_app.login_manager.unauthorized()
+
+        # flask 1.x compatibility
+        # current_app.ensure_sync is only available in Flask >= 2.0
+        if callable(getattr(current_app, "ensure_sync", None)):
+            return current_app.ensure_sync(func)(*args, **kwargs)
+        return func(*args, **kwargs)
+
+    return decorated_view
+
+
+def _get_user():
+    if has_request_context():
+        if "_login_user" not in g:
+            current_app.login_manager._load_user()
+
+        return g._login_user
+
+    return None

api/core/model_providers/model_provider_factory.py

@@ -57,6 +57,12 @@ class ModelProviderFactory:
         elif provider_name == 'huggingface_hub':
             from core.model_providers.providers.huggingface_hub_provider import HuggingfaceHubProvider
             return HuggingfaceHubProvider
+        elif provider_name == 'xinference':
+            from core.model_providers.providers.xinference_provider import XinferenceProvider
+            return XinferenceProvider
+        elif provider_name == 'openllm':
+            from core.model_providers.providers.openllm_provider import OpenLLMProvider
+            return OpenLLMProvider
         else:
             raise NotImplementedError
 

api/core/model_providers/models/embedding/xinference_embedding.py

@@ -0,0 +1,27 @@
+from core.third_party.langchain.embeddings.xinference_embedding import XinferenceEmbedding as XinferenceEmbeddings
+from replicate.exceptions import ModelError, ReplicateError
+
+from core.model_providers.error import LLMBadRequestError
+from core.model_providers.providers.base import BaseModelProvider
+from core.model_providers.models.embedding.base import BaseEmbedding
+
+
+class XinferenceEmbedding(BaseEmbedding):
+    def __init__(self, model_provider: BaseModelProvider, name: str):
+        credentials = model_provider.get_model_credentials(
+            model_name=name,
+            model_type=self.type
+        )
+
+        client = XinferenceEmbeddings(
+            server_url=credentials['server_url'],
+            model_uid=credentials['model_uid'],
+        )
+
+        super().__init__(model_provider, client, name)
+
+    def handle_exceptions(self, ex: Exception) -> Exception:
+        if isinstance(ex, (ModelError, ReplicateError)):
+            return LLMBadRequestError(f"Xinference embedding: {str(ex)}")
+        else:
+            return ex

api/core/model_providers/models/llm/openllm_model.py

@@ -0,0 +1,60 @@
+from typing import List, Optional, Any
+
+from langchain.callbacks.manager import Callbacks
+from langchain.schema import LLMResult
+
+from core.model_providers.error import LLMBadRequestError
+from core.model_providers.models.llm.base import BaseLLM
+from core.model_providers.models.entity.message import PromptMessage
+from core.model_providers.models.entity.model_params import ModelMode, ModelKwargs
+from core.third_party.langchain.llms.openllm import OpenLLM
+
+
+class OpenLLMModel(BaseLLM):
+    model_mode: ModelMode = ModelMode.COMPLETION
+
+    def _init_client(self) -> Any:
+        self.provider_model_kwargs = self._to_model_kwargs_input(self.model_rules, self.model_kwargs)
+
+        client = OpenLLM(
+            server_url=self.credentials.get('server_url'),
+            callbacks=self.callbacks,
+            llm_kwargs=self.provider_model_kwargs
+        )
+
+        return client
+
+    def _run(self, messages: List[PromptMessage],
+             stop: Optional[List[str]] = None,
+             callbacks: Callbacks = None,
+             **kwargs) -> LLMResult:
+        """
+        run predict by prompt messages and stop words.
+
+        :param messages:
+        :param stop:
+        :param callbacks:
+        :return:
+        """
+        prompts = self._get_prompt_from_messages(messages)
+        return self._client.generate([prompts], stop, callbacks)
+
+    def get_num_tokens(self, messages: List[PromptMessage]) -> int:
+        """
+        get num tokens of prompt messages.
+
+        :param messages:
+        :return:
+        """
+        prompts = self._get_prompt_from_messages(messages)
+        return max(self._client.get_num_tokens(prompts), 0)
+
+    def _set_model_kwargs(self, model_kwargs: ModelKwargs):
+        pass
+
+    def handle_exceptions(self, ex: Exception) -> Exception:
+        return LLMBadRequestError(f"OpenLLM: {str(ex)}")
+
+    @classmethod
+    def support_streaming(cls):
+        return False

api/core/model_providers/models/llm/xinference_model.py

@@ -0,0 +1,70 @@
+from typing import List, Optional, Any
+
+from langchain.callbacks.manager import Callbacks
+from langchain.schema import LLMResult
+
+from core.model_providers.error import LLMBadRequestError
+from core.model_providers.models.llm.base import BaseLLM
+from core.model_providers.models.entity.message import PromptMessage
+from core.model_providers.models.entity.model_params import ModelMode, ModelKwargs
+from core.third_party.langchain.llms.xinference_llm import XinferenceLLM
+
+
+class XinferenceModel(BaseLLM):
+    model_mode: ModelMode = ModelMode.COMPLETION
+
+    def _init_client(self) -> Any:
+        self.provider_model_kwargs = self._to_model_kwargs_input(self.model_rules, self.model_kwargs)
+
+        client = XinferenceLLM(
+            server_url=self.credentials['server_url'],
+            model_uid=self.credentials['model_uid'],
+        )
+
+        client.callbacks = self.callbacks
+
+        return client
+
+    def _run(self, messages: List[PromptMessage],
+             stop: Optional[List[str]] = None,
+             callbacks: Callbacks = None,
+             **kwargs) -> LLMResult:
+        """
+        run predict by prompt messages and stop words.
+
+        :param messages:
+        :param stop:
+        :param callbacks:
+        :return:
+        """
+        prompts = self._get_prompt_from_messages(messages)
+        return self._client.generate(
+            [prompts],
+            stop,
+            callbacks,
+            generate_config={
+                "stop": stop,
+                "stream": self.streaming,
+                **self.provider_model_kwargs,
+            }
+        )
+
+    def get_num_tokens(self, messages: List[PromptMessage]) -> int:
+        """
+        get num tokens of prompt messages.
+
+        :param messages:
+        :return:
+        """
+        prompts = self._get_prompt_from_messages(messages)
+        return max(self._client.get_num_tokens(prompts), 0)
+
+    def _set_model_kwargs(self, model_kwargs: ModelKwargs):
+        pass
+
+    def handle_exceptions(self, ex: Exception) -> Exception:
+        return LLMBadRequestError(f"Xinference: {str(ex)}")
+
+    @classmethod
+    def support_streaming(cls):
+        return True

api/core/model_providers/providers/openllm_provider.py

@@ -0,0 +1,138 @@
+import json
+from typing import Type
+
+from core.helper import encrypter
+from core.model_providers.models.entity.model_params import KwargRule, ModelKwargsRules, ModelType
+from core.model_providers.models.llm.openllm_model import OpenLLMModel
+from core.model_providers.providers.base import BaseModelProvider, CredentialsValidateFailedError
+
+from core.model_providers.models.base import BaseProviderModel
+from core.third_party.langchain.llms.openllm import OpenLLM
+from models.provider import ProviderType
+
+
+class OpenLLMProvider(BaseModelProvider):
+    @property
+    def provider_name(self):
+        """
+        Returns the name of a provider.
+        """
+        return 'openllm'
+
+    def _get_fixed_model_list(self, model_type: ModelType) -> list[dict]:
+        return []
+
+    def get_model_class(self, model_type: ModelType) -> Type[BaseProviderModel]:
+        """
+        Returns the model class.
+
+        :param model_type:
+        :return:
+        """
+        if model_type == ModelType.TEXT_GENERATION:
+            model_class = OpenLLMModel
+        else:
+            raise NotImplementedError
+
+        return model_class
+
+    def get_model_parameter_rules(self, model_name: str, model_type: ModelType) -> ModelKwargsRules:
+        """
+        get model parameter rules.
+
+        :param model_name:
+        :param model_type:
+        :return:
+        """
+        return ModelKwargsRules(
+            temperature=KwargRule[float](min=0.01, max=2, default=1),
+            top_p=KwargRule[float](min=0, max=1, default=0.7),
+            presence_penalty=KwargRule[float](min=-2, max=2, default=0),
+            frequency_penalty=KwargRule[float](min=-2, max=2, default=0),
+            max_tokens=KwargRule[int](alias='max_new_tokens', min=10, max=4000, default=128),
+        )
+
+    @classmethod
+    def is_model_credentials_valid_or_raise(cls, model_name: str, model_type: ModelType, credentials: dict):
+        """
+        check model credentials valid.
+
+        :param model_name:
+        :param model_type:
+        :param credentials:
+        """
+        if 'server_url' not in credentials:
+            raise CredentialsValidateFailedError('OpenLLM Server URL must be provided.')
+
+        try:
+            credential_kwargs = {
+                'server_url': credentials['server_url']
+            }
+
+            llm = OpenLLM(
+                llm_kwargs={
+                    'max_new_tokens': 10
+                },
+                **credential_kwargs
+            )
+
+            llm("ping")
+        except Exception as ex:
+            raise CredentialsValidateFailedError(str(ex))
+
+    @classmethod
+    def encrypt_model_credentials(cls, tenant_id: str, model_name: str, model_type: ModelType,
+                                  credentials: dict) -> dict:
+        """
+        encrypt model credentials for save.
+
+        :param tenant_id:
+        :param model_name:
+        :param model_type:
+        :param credentials:
+        :return:
+        """
+        credentials['server_url'] = encrypter.encrypt_token(tenant_id, credentials['server_url'])
+        return credentials
+
+    def get_model_credentials(self, model_name: str, model_type: ModelType, obfuscated: bool = False) -> dict:
+        """
+        get credentials for llm use.
+
+        :param model_name:
+        :param model_type:
+        :param obfuscated:
+        :return:
+        """
+        if self.provider.provider_type != ProviderType.CUSTOM.value:
+            raise NotImplementedError
+
+        provider_model = self._get_provider_model(model_name, model_type)
+
+        if not provider_model.encrypted_config:
+            return {
+                'server_url': None
+            }
+
+        credentials = json.loads(provider_model.encrypted_config)
+        if credentials['server_url']:
+            credentials['server_url'] = encrypter.decrypt_token(
+                self.provider.tenant_id,
+                credentials['server_url']
+            )
+
+            if obfuscated:
+                credentials['server_url'] = encrypter.obfuscated_token(credentials['server_url'])
+
+        return credentials
+
+    @classmethod
+    def is_provider_credentials_valid_or_raise(cls, credentials: dict):
+        return
+
+    @classmethod
+    def encrypt_provider_credentials(cls, tenant_id: str, credentials: dict) -> dict:
+        return {}
+
+    def get_provider_credentials(self, obfuscated: bool = False) -> dict:
+        return {}

api/core/model_providers/providers/xinference_provider.py

@@ -0,0 +1,193 @@
+import json
+from typing import Type
+
+import requests
+from xinference.client import RESTfulGenerateModelHandle, RESTfulChatModelHandle, RESTfulChatglmCppChatModelHandle
+
+from core.helper import encrypter
+from core.model_providers.models.embedding.xinference_embedding import XinferenceEmbedding
+from core.model_providers.models.entity.model_params import KwargRule, ModelKwargsRules, ModelType
+from core.model_providers.models.llm.xinference_model import XinferenceModel
+from core.model_providers.providers.base import BaseModelProvider, CredentialsValidateFailedError
+
+from core.model_providers.models.base import BaseProviderModel
+from core.third_party.langchain.llms.xinference_llm import XinferenceLLM
+from models.provider import ProviderType
+
+
+class XinferenceProvider(BaseModelProvider):
+    @property
+    def provider_name(self):
+        """
+        Returns the name of a provider.
+        """
+        return 'xinference'
+
+    def _get_fixed_model_list(self, model_type: ModelType) -> list[dict]:
+        return []
+
+    def get_model_class(self, model_type: ModelType) -> Type[BaseProviderModel]:
+        """
+        Returns the model class.
+
+        :param model_type:
+        :return:
+        """
+        if model_type == ModelType.TEXT_GENERATION:
+            model_class = XinferenceModel
+        elif model_type == ModelType.EMBEDDINGS:
+            model_class = XinferenceEmbedding
+        else:
+            raise NotImplementedError
+
+        return model_class
+
+    def get_model_parameter_rules(self, model_name: str, model_type: ModelType) -> ModelKwargsRules:
+        """
+        get model parameter rules.
+
+        :param model_name:
+        :param model_type:
+        :return:
+        """
+        credentials = self.get_model_credentials(model_name, model_type)
+        if credentials['model_format'] == "ggmlv3" and credentials["model_handle_type"] == "chatglm":
+            return ModelKwargsRules(
+                temperature=KwargRule[float](min=0.01, max=2, default=1),
+                top_p=KwargRule[float](min=0, max=1, default=0.7),
+                presence_penalty=KwargRule[float](enabled=False),
+                frequency_penalty=KwargRule[float](enabled=False),
+                max_tokens=KwargRule[int](min=10, max=4000, default=256),
+            )
+        elif credentials['model_format'] == "ggmlv3":
+            return ModelKwargsRules(
+                temperature=KwargRule[float](min=0.01, max=2, default=1),
+                top_p=KwargRule[float](min=0, max=1, default=0.7),
+                presence_penalty=KwargRule[float](min=-2, max=2, default=0),
+                frequency_penalty=KwargRule[float](min=-2, max=2, default=0),
+                max_tokens=KwargRule[int](min=10, max=4000, default=256),
+            )
+        else:
+            return ModelKwargsRules(
+                temperature=KwargRule[float](min=0.01, max=2, default=1),
+                top_p=KwargRule[float](min=0, max=1, default=0.7),
+                presence_penalty=KwargRule[float](enabled=False),
+                frequency_penalty=KwargRule[float](enabled=False),
+                max_tokens=KwargRule[int](alias='max_new_tokens', min=10, max=4000, default=256),
+            )
+
+
+    @classmethod
+    def is_model_credentials_valid_or_raise(cls, model_name: str, model_type: ModelType, credentials: dict):
+        """
+        check model credentials valid.
+
+        :param model_name:
+        :param model_type:
+        :param credentials:
+        """
+        if 'server_url' not in credentials:
+            raise CredentialsValidateFailedError('Xinference Server URL must be provided.')
+
+        if 'model_uid' not in credentials:
+            raise CredentialsValidateFailedError('Xinference Model UID must be provided.')
+
+        try:
+            credential_kwargs = {
+                'server_url': credentials['server_url'],
+                'model_uid': credentials['model_uid'],
+            }
+
+            llm = XinferenceLLM(
+                **credential_kwargs
+            )
+
+            llm("ping")
+        except Exception as ex:
+            raise CredentialsValidateFailedError(str(ex))
+
+    @classmethod
+    def encrypt_model_credentials(cls, tenant_id: str, model_name: str, model_type: ModelType,
+                                  credentials: dict) -> dict:
+        """
+        encrypt model credentials for save.
+
+        :param tenant_id:
+        :param model_name:
+        :param model_type:
+        :param credentials:
+        :return:
+        """
+        extra_credentials = cls._get_extra_credentials(credentials)
+        credentials.update(extra_credentials)
+
+        credentials['server_url'] = encrypter.encrypt_token(tenant_id, credentials['server_url'])
+
+        return credentials
+
+    def get_model_credentials(self, model_name: str, model_type: ModelType, obfuscated: bool = False) -> dict:
+        """
+        get credentials for llm use.
+
+        :param model_name:
+        :param model_type:
+        :param obfuscated:
+        :return:
+        """
+        if self.provider.provider_type != ProviderType.CUSTOM.value:
+            raise NotImplementedError
+
+        provider_model = self._get_provider_model(model_name, model_type)
+
+        if not provider_model.encrypted_config:
+            return {
+                'server_url': None,
+                'model_uid': None,
+            }
+
+        credentials = json.loads(provider_model.encrypted_config)
+        if credentials['server_url']:
+            credentials['server_url'] = encrypter.decrypt_token(
+                self.provider.tenant_id,
+                credentials['server_url']
+            )
+
+            if obfuscated:
+                credentials['server_url'] = encrypter.obfuscated_token(credentials['server_url'])
+
+        return credentials
+
+    @classmethod
+    def _get_extra_credentials(self, credentials: dict) -> dict:
+        url = f"{credentials['server_url']}/v1/models/{credentials['model_uid']}"
+        response = requests.get(url)
+        if response.status_code != 200:
+            raise RuntimeError(
+                f"Failed to get the model description, detail: {response.json()['detail']}"
+            )
+        desc = response.json()
+
+        extra_credentials = {
+            'model_format': desc['model_format'],
+        }
+        if desc["model_format"] == "ggmlv3" and "chatglm" in desc["model_name"]:
+            extra_credentials['model_handle_type'] = 'chatglm'
+        elif "generate" in desc["model_ability"]:
+            extra_credentials['model_handle_type'] = 'generate'
+        elif "chat" in desc["model_ability"]:
+            extra_credentials['model_handle_type'] = 'chat'
+        else:
+            raise NotImplementedError(f"Model handle type not supported.")
+
+        return extra_credentials
+
+    @classmethod
+    def is_provider_credentials_valid_or_raise(cls, credentials: dict):
+        return
+
+    @classmethod
+    def encrypt_provider_credentials(cls, tenant_id: str, credentials: dict) -> dict:
+        return {}
+
+    def get_provider_credentials(self, obfuscated: bool = False) -> dict:
+        return {}

api/core/model_providers/rules/_providers.json

@@ -8,5 +8,7 @@
   "wenxin",
   "chatglm",
   "replicate",
-  "huggingface_hub"
+  "huggingface_hub",
+  "xinference",
+  "openllm"
 ]

api/core/model_providers/rules/openllm.json

@@ -0,0 +1,7 @@
+{
+    "support_provider_types": [
+        "custom"
+    ],
+    "system_config": null,
+    "model_flexibility": "configurable"
+}

api/core/model_providers/rules/xinference.json

@@ -0,0 +1,7 @@
+{
+    "support_provider_types": [
+        "custom"
+    ],
+    "system_config": null,
+    "model_flexibility": "configurable"
+}

api/core/third_party/langchain/embeddings/xinference_embedding.py

@@ -0,0 +1,21 @@
+from typing import List
+
+import numpy as np
+from langchain.embeddings import XinferenceEmbeddings
+
+
+class XinferenceEmbedding(XinferenceEmbeddings):
+
+    def embed_documents(self, texts: List[str]) -> List[List[float]]:
+        vectors = super().embed_documents(texts)
+
+        normalized_vectors = [(vector / np.linalg.norm(vector)).tolist() for vector in vectors]
+
+        return normalized_vectors
+
+    def embed_query(self, text: str) -> List[float]:
+        vector = super().embed_query(text)
+
+        normalized_vector = (vector / np.linalg.norm(vector)).tolist()
+
+        return normalized_vector

api/core/third_party/langchain/llms/openllm.py

@@ -0,0 +1,84 @@
+from __future__ import annotations
+
+import logging
+from typing import (
+    Any,
+    Dict,
+    List,
+    Optional,
+)
+
+import requests
+from langchain.llms.utils import enforce_stop_tokens
+from pydantic import Field
+
+from langchain.callbacks.manager import (
+    AsyncCallbackManagerForLLMRun,
+    CallbackManagerForLLMRun,
+)
+from langchain.llms.base import LLM
+
+logger = logging.getLogger(__name__)
+
+
+class OpenLLM(LLM):
+    """OpenLLM, supporting both in-process model
+    instance and remote OpenLLM servers.
+
+    If you have a OpenLLM server running, you can also use it remotely:
+        .. code-block:: python
+
+            from langchain.llms import OpenLLM
+            llm = OpenLLM(server_url='http://localhost:3000')
+            llm("What is the difference between a duck and a goose?")
+    """
+
+    server_url: Optional[str] = None
+    """Optional server URL that currently runs a LLMServer with 'openllm start'."""
+    llm_kwargs: Dict[str, Any] = Field(default_factory=dict)
+    """Key word arguments to be passed to openllm.LLM"""
+
+    @property
+    def _llm_type(self) -> str:
+        return "openllm"
+
+    def _call(
+        self,
+        prompt: str,
+        stop: Optional[List[str]] = None,
+        run_manager: CallbackManagerForLLMRun | None = None,
+        **kwargs: Any,
+    ) -> str:
+        params = {
+            "prompt": prompt,
+            "llm_config": self.llm_kwargs
+        }
+
+        headers = {"Content-Type": "application/json"}
+        response = requests.post(
+            f'{self.server_url}/v1/generate',
+            headers=headers,
+            json=params
+        )
+
+        if not response.ok:
+            raise ValueError(f"OpenLLM HTTP {response.status_code} error: {response.text}")
+
+        json_response = response.json()
+        completion = json_response["responses"][0]
+
+        if stop is not None:
+            completion = enforce_stop_tokens(completion, stop)
+
+        return completion
+
+    async def _acall(
+        self,
+        prompt: str,
+        stop: Optional[List[str]] = None,
+        run_manager: Optional[AsyncCallbackManagerForLLMRun] = None,
+        **kwargs: Any,
+    ) -> str:
+        raise NotImplementedError(
+            "Async call is not supported for OpenLLM at the moment."
+        )

api/core/third_party/langchain/llms/xinference_llm.py

@@ -0,0 +1,132 @@
+from typing import Optional, List, Any, Union, Generator
+
+from langchain.callbacks.manager import CallbackManagerForLLMRun
+from langchain.llms import Xinference
+from langchain.llms.utils import enforce_stop_tokens
+from xinference.client import RESTfulChatglmCppChatModelHandle, \
+    RESTfulChatModelHandle, RESTfulGenerateModelHandle
+
+
+class XinferenceLLM(Xinference):
+    def _call(
+        self,
+        prompt: str,
+        stop: Optional[List[str]] = None,
+        run_manager: Optional[CallbackManagerForLLMRun] = None,
+        **kwargs: Any,
+    ) -> str:
+        """Call the xinference model and return the output.
+
+        Args:
+            prompt: The prompt to use for generation.
+            stop: Optional list of stop words to use when generating.
+            generate_config: Optional dictionary for the configuration used for
+                generation.
+
+        Returns:
+            The generated string by the model.
+        """
+        model = self.client.get_model(self.model_uid)
+
+        if isinstance(model, RESTfulChatModelHandle):
+            generate_config: "LlamaCppGenerateConfig" = kwargs.get("generate_config", {})
+
+            if stop:
+                generate_config["stop"] = stop
+
+            if generate_config and generate_config.get("stream"):
+                combined_text_output = ""
+                for token in self._stream_generate(
+                        model=model,
+                        prompt=prompt,
+                        run_manager=run_manager,
+                        generate_config=generate_config,
+                ):
+                    combined_text_output += token
+                return combined_text_output
+            else:
+                completion = model.chat(prompt=prompt, generate_config=generate_config)
+                return completion["choices"][0]["message"]["content"]
+        elif isinstance(model, RESTfulGenerateModelHandle):
+            generate_config: "LlamaCppGenerateConfig" = kwargs.get("generate_config", {})
+
+            if stop:
+                generate_config["stop"] = stop
+
+            if generate_config and generate_config.get("stream"):
+                combined_text_output = ""
+                for token in self._stream_generate(
+                    model=model,
+                    prompt=prompt,
+                    run_manager=run_manager,
+                    generate_config=generate_config,
+                ):
+                    combined_text_output += token
+                return combined_text_output
+
+            else:
+                completion = model.generate(prompt=prompt, generate_config=generate_config)
+                return completion["choices"][0]["text"]
+        elif isinstance(model, RESTfulChatglmCppChatModelHandle):
+            generate_config: "ChatglmCppGenerateConfig" = kwargs.get("generate_config", {})
+
+            if generate_config and generate_config.get("stream"):
+                combined_text_output = ""
+                for token in self._stream_generate(
+                    model=model,
+                    prompt=prompt,
+                    run_manager=run_manager,
+                    generate_config=generate_config,
+                ):
+                    combined_text_output += token
+                completion = combined_text_output
+            else:
+                completion = model.chat(prompt=prompt, generate_config=generate_config)
+                completion = completion["choices"][0]["message"]["content"]
+
+            if stop is not None:
+                completion = enforce_stop_tokens(completion, stop)
+
+            return completion
+
+
+    def _stream_generate(
+        self,
+        model: Union["RESTfulGenerateModelHandle", "RESTfulChatModelHandle", "RESTfulChatglmCppChatModelHandle"],
+        prompt: str,
+        run_manager: Optional[CallbackManagerForLLMRun] = None,
+        generate_config: Optional[Union["LlamaCppGenerateConfig", "PytorchGenerateConfig", "ChatglmCppGenerateConfig"]] = None,
+    ) -> Generator[str, None, None]:
+        """
+        Args:
+            prompt: The prompt to use for generation.
+            model: The model used for generation.
+            stop: Optional list of stop words to use when generating.
+            generate_config: Optional dictionary for the configuration used for
+                generation.
+
+        Yields:
+            A string token.
+        """
+        if isinstance(model, RESTfulGenerateModelHandle):
+            streaming_response = model.generate(
+                prompt=prompt, generate_config=generate_config
+            )
+        else:
+            streaming_response = model.chat(
+                prompt=prompt, generate_config=generate_config
+            )
+
+        for chunk in streaming_response:
+            if isinstance(chunk, dict):
+                choices = chunk.get("choices", [])
+                if choices:
+                    choice = choices[0]
+                    if isinstance(choice, dict):
+                        token = choice.get("text", "")
+                        log_probs = choice.get("logprobs")
+                        if run_manager:
+                            run_manager.on_llm_new_token(
+                                token=token, verbose=self.verbose, log_probs=log_probs
+                            )
+                        yield token

api/core/tool/web_reader_tool.py

@@ -88,6 +88,11 @@ class WebReaderTool(BaseTool):
             texts = character_splitter.split_text(page_contents)
             docs = [Document(page_content=t) for t in texts]
 
+            if len(docs) == 0:
+                return "No content found."
+
+            docs = docs[1:]
+
             # only use first 5 docs
             if len(docs) > 5:
                 docs = docs[:5]

api/events/event_handlers/generate_conversation_name_when_first_message_created.py

@@ -20,6 +20,10 @@ def handle(sender, **kwargs):
             # generate conversation name
             try:
                 name = LLMGenerator.generate_conversation_name(app_model.tenant_id, message.query, message.answer)
+
+                if len(name) > 75:
+                    name = name[:75] + '...'
+
                 conversation.name = name
             except:
                 conversation.name = 'New Chat'

api/requirements.txt

@@ -48,4 +48,5 @@ dashscope~=1.5.0
 huggingface_hub~=0.16.4
 transformers~=4.31.0
 stripe~=5.5.0
-pandas==1.5.3
+pandas==1.5.3
+xinference==0.2.1

api/services/dataset_service.py

@@ -284,8 +284,9 @@ class DocumentService:
             "github_link": str,
             "open_source_license": str,
             "commit_date": str,
-            "commit_author": str
-        }
+            "commit_author": str,
+        },
+        "others": dict
     }
 
     @staticmethod
@@ -370,8 +371,8 @@ class DocumentService:
             raise DocumentIndexingError()
         # update document to be recover
         document.is_paused = False
-        document.paused_by = current_user.id
-        document.paused_at = time.time()
+        document.paused_by = None
+        document.paused_at = None
 
         db.session.add(document)
         db.session.commit()
@@ -972,7 +973,7 @@ class SegmentService:
                 db.session.add(segment)
                 db.session.commit()
                 # update segment vector index
-                VectorService.create_segment_vector(args['keywords'], segment, dataset)
+                VectorService.update_segment_vector(args['keywords'], segment, dataset)
         except Exception as e:
             logging.exception("update segment index failed")
             segment.enabled = False

api/tests/integration_tests/.env.example

@@ -32,4 +32,11 @@ WENXIN_API_KEY=
 WENXIN_SECRET_KEY=
 
 # ChatGLM Credentials
-CHATGLM_API_BASE=
+CHATGLM_API_BASE=
+
+# Xinference Credentials
+XINFERENCE_SERVER_URL=
+XINFERENCE_MODEL_UID=
+
+# OpenLLM Credentials
+OPENLLM_SERVER_URL=

api/tests/integration_tests/models/embedding/test_xinference_embedding.py

@@ -0,0 +1,65 @@
+import json
+import os
+from unittest.mock import patch, MagicMock
+
+from core.model_providers.models.embedding.xinference_embedding import XinferenceEmbedding
+from core.model_providers.models.entity.model_params import ModelType
+from core.model_providers.providers.xinference_provider import XinferenceProvider
+from models.provider import Provider, ProviderType, ProviderModel
+
+
+def get_mock_provider():
+    return Provider(
+        id='provider_id',
+        tenant_id='tenant_id',
+        provider_name='xinference',
+        provider_type=ProviderType.CUSTOM.value,
+        encrypted_config='',
+        is_valid=True,
+    )
+
+
+def get_mock_embedding_model(mocker):
+    model_name = 'vicuna-v1.3'
+    server_url = os.environ['XINFERENCE_SERVER_URL']
+    model_uid = os.environ['XINFERENCE_MODEL_UID']
+    model_provider = XinferenceProvider(provider=get_mock_provider())
+
+    mock_query = MagicMock()
+    mock_query.filter.return_value.first.return_value = ProviderModel(
+        provider_name='xinference',
+        model_name=model_name,
+        model_type=ModelType.EMBEDDINGS.value,
+        encrypted_config=json.dumps({
+            'server_url': server_url,
+            'model_uid': model_uid
+        }),
+        is_valid=True,
+    )
+    mocker.patch('extensions.ext_database.db.session.query', return_value=mock_query)
+
+    return XinferenceEmbedding(
+        model_provider=model_provider,
+        name=model_name
+    )
+
+
+def decrypt_side_effect(tenant_id, encrypted_api_key):
+    return encrypted_api_key
+
+
+@patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
+def test_embed_documents(mock_decrypt, mocker):
+    embedding_model = get_mock_embedding_model(mocker)
+    rst = embedding_model.client.embed_documents(['test', 'test1'])
+    assert isinstance(rst, list)
+    assert len(rst) == 2
+    assert len(rst[0]) == 4096
+
+
+@patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
+def test_embed_query(mock_decrypt, mocker):
+    embedding_model = get_mock_embedding_model(mocker)
+    rst = embedding_model.client.embed_query('test')
+    assert isinstance(rst, list)
+    assert len(rst) == 4096

api/tests/integration_tests/models/llm/test_anthropic_model.py

@@ -50,7 +50,9 @@ def test_get_num_tokens(mock_decrypt):
 
 
 @patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
-def test_run(mock_decrypt):
+def test_run(mock_decrypt, mocker):
+    mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
+
     model = get_mock_model('claude-2')
     messages = [PromptMessage(content='Human: 1 + 1=? \nAssistant: ')]
     rst = model.run(
@@ -58,4 +60,3 @@ def test_run(mock_decrypt):
         stop=['\nHuman:'],
     )
     assert len(rst.content) > 0
-    assert rst.content.strip() == '2'

api/tests/integration_tests/models/llm/test_azure_openai_model.py

@@ -76,6 +76,8 @@ def test_chat_get_num_tokens(mock_decrypt, mocker):
 
 @patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
 def test_run(mock_decrypt, mocker):
+    mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
+
     openai_model = get_mock_azure_openai_model('gpt-35-turbo', mocker)
     messages = [PromptMessage(content='Human: Are you Human? you MUST only answer `y` or `n`? \nAssistant: ')]
     rst = openai_model.run(
@@ -83,4 +85,3 @@ def test_run(mock_decrypt, mocker):
         stop=['\nHuman:'],
     )
     assert len(rst.content) > 0
-    assert rst.content.strip() == 'n'

api/tests/integration_tests/models/llm/test_huggingface_hub_model.py

@@ -95,6 +95,8 @@ def test_inference_endpoints_get_num_tokens(mock_decrypt, mock_model_info, mocke
 
 @patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
 def test_hosted_inference_api_run(mock_decrypt, mocker):
+    mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
+
     model = get_mock_model(
         'google/flan-t5-base',
         'hosted_inference_api',
@@ -111,6 +113,8 @@ def test_hosted_inference_api_run(mock_decrypt, mocker):
 
 @patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
 def test_inference_endpoints_run(mock_decrypt, mocker):
+    mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
+
     model = get_mock_model(
         '',
         'inference_endpoints',
@@ -121,4 +125,3 @@ def test_inference_endpoints_run(mock_decrypt, mocker):
         [PromptMessage(content='Answer the following yes/no question. Can you write a whole Haiku in a single tweet?')],
     )
     assert len(rst.content) > 0
-    assert rst.content.strip() == 'no'

api/tests/integration_tests/models/llm/test_minimax_model.py

@@ -54,11 +54,12 @@ def test_get_num_tokens(mock_decrypt):
 
 
 @patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
-def test_run(mock_decrypt):
+def test_run(mock_decrypt, mocker):
+    mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
+
     model = get_mock_model('abab5.5-chat')
     rst = model.run(
         [PromptMessage(content='Human: Are you a real Human? you MUST only answer `y` or `n`? \nAssistant: ')],
         stop=['\nHuman:'],
     )
     assert len(rst.content) > 0
-    assert rst.content.strip() == 'n'

api/tests/integration_tests/models/llm/test_openai_model.py

@@ -58,7 +58,9 @@ def test_chat_get_num_tokens(mock_decrypt):
 
 
 @patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
-def test_run(mock_decrypt):
+def test_run(mock_decrypt, mocker):
+    mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
+
     openai_model = get_mock_openai_model('text-davinci-003')
     rst = openai_model.run(
         [PromptMessage(content='Human: Are you Human? you MUST only answer `y` or `n`? \nAssistant: ')],
@@ -69,7 +71,9 @@ def test_run(mock_decrypt):
 
 
 @patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
-def test_chat_run(mock_decrypt):
+def test_chat_run(mock_decrypt, mocker):
+    mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
+
     openai_model = get_mock_openai_model('gpt-3.5-turbo')
     messages = [PromptMessage(content='Human: Are you Human? you MUST only answer `y` or `n`? \nAssistant: ')]
     rst = openai_model.run(
@@ -77,4 +81,3 @@ def test_chat_run(mock_decrypt):
         stop=['\nHuman:'],
     )
     assert len(rst.content) > 0
-    assert rst.content.strip() == 'n'

api/tests/integration_tests/models/llm/test_openllm_model.py

@@ -0,0 +1,72 @@
+import json
+import os
+from unittest.mock import patch, MagicMock
+
+from core.model_providers.models.entity.message import PromptMessage, MessageType
+from core.model_providers.models.entity.model_params import ModelKwargs, ModelType
+from core.model_providers.models.llm.openllm_model import OpenLLMModel
+from core.model_providers.providers.openllm_provider import OpenLLMProvider
+from models.provider import Provider, ProviderType, ProviderModel
+
+
+def get_mock_provider():
+    return Provider(
+        id='provider_id',
+        tenant_id='tenant_id',
+        provider_name='openllm',
+        provider_type=ProviderType.CUSTOM.value,
+        encrypted_config='',
+        is_valid=True,
+    )
+
+
+def get_mock_model(model_name, mocker):
+    model_kwargs = ModelKwargs(
+        max_tokens=10,
+        temperature=0.01
+    )
+    server_url = os.environ['OPENLLM_SERVER_URL']
+    model_provider = OpenLLMProvider(provider=get_mock_provider())
+
+    mock_query = MagicMock()
+    mock_query.filter.return_value.first.return_value = ProviderModel(
+        provider_name='openllm',
+        model_name=model_name,
+        model_type=ModelType.TEXT_GENERATION.value,
+        encrypted_config=json.dumps({
+            'server_url': server_url
+        }),
+        is_valid=True,
+    )
+    mocker.patch('extensions.ext_database.db.session.query', return_value=mock_query)
+
+    return OpenLLMModel(
+        model_provider=model_provider,
+        name=model_name,
+        model_kwargs=model_kwargs
+    )
+
+
+def decrypt_side_effect(tenant_id, encrypted_api_key):
+    return encrypted_api_key
+
+
+@patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
+def test_get_num_tokens(mock_decrypt, mocker):
+    model = get_mock_model('facebook/opt-125m', mocker)
+    rst = model.get_num_tokens([
+        PromptMessage(type=MessageType.HUMAN, content='Who is your manufacturer?')
+    ])
+    assert rst == 5
+
+
+@patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
+def test_run(mock_decrypt, mocker):
+    mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
+
+    model = get_mock_model('facebook/opt-125m', mocker)
+    messages = [PromptMessage(content='Human: who are you? \nAnswer: ')]
+    rst = model.run(
+        messages
+    )
+    assert len(rst.content) > 0

api/tests/integration_tests/models/llm/test_replicate_model.py

@@ -65,6 +65,8 @@ def test_get_num_tokens(mock_decrypt, mocker):
 
 @patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
 def test_run(mock_decrypt, mocker):
+    mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
+
     model = get_mock_model('a16z-infra/llama-2-13b-chat', '2a7f981751ec7fdf87b5b91ad4db53683a98082e9ff7bfd12c8cd5ea85980a52', mocker)
     messages = [PromptMessage(content='Human: 1+1=? \nAnswer: ')]
     rst = model.run(

api/tests/integration_tests/models/llm/test_spark_model.py

@@ -58,7 +58,9 @@ def test_get_num_tokens(mock_decrypt):
 
 
 @patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
-def test_run(mock_decrypt):
+def test_run(mock_decrypt, mocker):
+    mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
+
     model = get_mock_model('spark')
     messages = [PromptMessage(content='Human: 1 + 1=? \nAssistant: Integer answer is:')]
     rst = model.run(
@@ -66,4 +68,3 @@ def test_run(mock_decrypt):
         stop=['\nHuman:'],
     )
     assert len(rst.content) > 0
-    assert rst.content.strip() == '2'

api/tests/integration_tests/models/llm/test_tongyi_model.py

@@ -52,7 +52,9 @@ def test_get_num_tokens(mock_decrypt):
 
 
 @patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
-def test_run(mock_decrypt):
+def test_run(mock_decrypt, mocker):
+    mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
+
     model = get_mock_model('qwen-v1')
     rst = model.run(
         [PromptMessage(content='Human: Are you Human? you MUST only answer `y` or `n`? \nAssistant: ')],

api/tests/integration_tests/models/llm/test_wenxin_model.py

@@ -52,7 +52,9 @@ def test_get_num_tokens(mock_decrypt):
 
 
 @patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
-def test_run(mock_decrypt):
+def test_run(mock_decrypt, mocker):
+    mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
+
     model = get_mock_model('ernie-bot')
     messages = [PromptMessage(content='Human: 1 + 1=? \nAssistant: Integer answer is:')]
     rst = model.run(
@@ -60,4 +62,3 @@ def test_run(mock_decrypt):
         stop=['\nHuman:'],
     )
     assert len(rst.content) > 0
-    assert rst.content.strip() == '2'

api/tests/integration_tests/models/llm/test_xinference_model.py

@@ -0,0 +1,74 @@
+import json
+import os
+from unittest.mock import patch, MagicMock
+
+from core.model_providers.models.entity.message import PromptMessage, MessageType
+from core.model_providers.models.entity.model_params import ModelKwargs, ModelType
+from core.model_providers.models.llm.xinference_model import XinferenceModel
+from core.model_providers.providers.xinference_provider import XinferenceProvider
+from models.provider import Provider, ProviderType, ProviderModel
+
+
+def get_mock_provider():
+    return Provider(
+        id='provider_id',
+        tenant_id='tenant_id',
+        provider_name='xinference',
+        provider_type=ProviderType.CUSTOM.value,
+        encrypted_config='',
+        is_valid=True,
+    )
+
+
+def get_mock_model(model_name, mocker):
+    model_kwargs = ModelKwargs(
+        max_tokens=10,
+        temperature=0.01
+    )
+    server_url = os.environ['XINFERENCE_SERVER_URL']
+    model_uid = os.environ['XINFERENCE_MODEL_UID']
+    model_provider = XinferenceProvider(provider=get_mock_provider())
+
+    mock_query = MagicMock()
+    mock_query.filter.return_value.first.return_value = ProviderModel(
+        provider_name='xinference',
+        model_name=model_name,
+        model_type=ModelType.TEXT_GENERATION.value,
+        encrypted_config=json.dumps({
+            'server_url': server_url,
+            'model_uid': model_uid
+        }),
+        is_valid=True,
+    )
+    mocker.patch('extensions.ext_database.db.session.query', return_value=mock_query)
+
+    return XinferenceModel(
+        model_provider=model_provider,
+        name=model_name,
+        model_kwargs=model_kwargs
+    )
+
+
+def decrypt_side_effect(tenant_id, encrypted_api_key):
+    return encrypted_api_key
+
+
+@patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
+def test_get_num_tokens(mock_decrypt, mocker):
+    model = get_mock_model('llama-2-chat', mocker)
+    rst = model.get_num_tokens([
+        PromptMessage(type=MessageType.HUMAN, content='Who is your manufacturer?')
+    ])
+    assert rst == 5
+
+
+@patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
+def test_run(mock_decrypt, mocker):
+    mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
+
+    model = get_mock_model('llama-2-chat', mocker)
+    messages = [PromptMessage(content='Human: 1+1=? \nAnswer: ')]
+    rst = model.run(
+        messages
+    )
+    assert len(rst.content) > 0

api/tests/unit_tests/model_providers/test_openllm_provider.py

@@ -0,0 +1,122 @@
+import pytest
+from unittest.mock import patch, MagicMock
+import json
+
+from core.model_providers.models.entity.model_params import ModelType
+from core.model_providers.providers.base import CredentialsValidateFailedError
+from core.model_providers.providers.openllm_provider import OpenLLMProvider
+from models.provider import ProviderType, Provider, ProviderModel
+
+PROVIDER_NAME = 'openllm'
+MODEL_PROVIDER_CLASS = OpenLLMProvider
+VALIDATE_CREDENTIAL = {
+    'server_url': 'http://127.0.0.1:3333/'
+}
+
+
+def encrypt_side_effect(tenant_id, encrypt_key):
+    return f'encrypted_{encrypt_key}'
+
+
+def decrypt_side_effect(tenant_id, encrypted_key):
+    return encrypted_key.replace('encrypted_', '')
+
+
+def test_is_credentials_valid_or_raise_valid(mocker):
+    mocker.patch('core.third_party.langchain.llms.openllm.OpenLLM._call',
+                 return_value="abc")
+
+    MODEL_PROVIDER_CLASS.is_model_credentials_valid_or_raise(
+        model_name='username/test_model_name',
+        model_type=ModelType.TEXT_GENERATION,
+        credentials=VALIDATE_CREDENTIAL.copy()
+    )
+
+
+def test_is_credentials_valid_or_raise_invalid(mocker):
+    # raise CredentialsValidateFailedError if credential is not in credentials
+    with pytest.raises(CredentialsValidateFailedError):
+        MODEL_PROVIDER_CLASS.is_model_credentials_valid_or_raise(
+            model_name='test_model_name',
+            model_type=ModelType.TEXT_GENERATION,
+            credentials={}
+        )
+
+    # raise CredentialsValidateFailedError if credential is invalid
+    with pytest.raises(CredentialsValidateFailedError):
+        MODEL_PROVIDER_CLASS.is_model_credentials_valid_or_raise(
+            model_name='test_model_name',
+            model_type=ModelType.TEXT_GENERATION,
+            credentials={'server_url': 'invalid'})
+
+
+@patch('core.helper.encrypter.encrypt_token', side_effect=encrypt_side_effect)
+def test_encrypt_model_credentials(mock_encrypt):
+    api_key = 'http://127.0.0.1:3333/'
+    result = MODEL_PROVIDER_CLASS.encrypt_model_credentials(
+        tenant_id='tenant_id',
+        model_name='test_model_name',
+        model_type=ModelType.TEXT_GENERATION,
+        credentials=VALIDATE_CREDENTIAL.copy()
+    )
+    mock_encrypt.assert_called_with('tenant_id', api_key)
+    assert result['server_url'] == f'encrypted_{api_key}'
+
+
+@patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
+def test_get_model_credentials_custom(mock_decrypt, mocker):
+    provider = Provider(
+        id='provider_id',
+        tenant_id='tenant_id',
+        provider_name=PROVIDER_NAME,
+        provider_type=ProviderType.CUSTOM.value,
+        encrypted_config=None,
+        is_valid=True,
+    )
+
+    encrypted_credential = VALIDATE_CREDENTIAL.copy()
+    encrypted_credential['server_url'] = 'encrypted_' + encrypted_credential['server_url']
+
+    mock_query = MagicMock()
+    mock_query.filter.return_value.first.return_value = ProviderModel(
+        encrypted_config=json.dumps(encrypted_credential)
+    )
+    mocker.patch('extensions.ext_database.db.session.query', return_value=mock_query)
+
+    model_provider = MODEL_PROVIDER_CLASS(provider=provider)
+    result = model_provider.get_model_credentials(
+        model_name='test_model_name',
+        model_type=ModelType.TEXT_GENERATION
+    )
+    assert result['server_url'] == 'http://127.0.0.1:3333/'
+
+
+@patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
+def test_get_model_credentials_obfuscated(mock_decrypt, mocker):
+    provider = Provider(
+        id='provider_id',
+        tenant_id='tenant_id',
+        provider_name=PROVIDER_NAME,
+        provider_type=ProviderType.CUSTOM.value,
+        encrypted_config=None,
+        is_valid=True,
+    )
+
+    encrypted_credential = VALIDATE_CREDENTIAL.copy()
+    encrypted_credential['server_url'] = 'encrypted_' + encrypted_credential['server_url']
+
+    mock_query = MagicMock()
+    mock_query.filter.return_value.first.return_value = ProviderModel(
+        encrypted_config=json.dumps(encrypted_credential)
+    )
+    mocker.patch('extensions.ext_database.db.session.query', return_value=mock_query)
+
+    model_provider = MODEL_PROVIDER_CLASS(provider=provider)
+    result = model_provider.get_model_credentials(
+        model_name='test_model_name',
+        model_type=ModelType.TEXT_GENERATION,
+        obfuscated=True
+    )
+    middle_token = result['server_url'][6:-2]
+    assert len(middle_token) == max(len(VALIDATE_CREDENTIAL['server_url']) - 8, 0)
+    assert all(char == '*' for char in middle_token)

api/tests/unit_tests/model_providers/test_xinference_provider.py

@@ -0,0 +1,130 @@
+import pytest
+from unittest.mock import patch, MagicMock
+import json
+
+from core.model_providers.models.entity.model_params import ModelType
+from core.model_providers.providers.base import CredentialsValidateFailedError
+from core.model_providers.providers.xinference_provider import XinferenceProvider
+from models.provider import ProviderType, Provider, ProviderModel
+
+PROVIDER_NAME = 'xinference'
+MODEL_PROVIDER_CLASS = XinferenceProvider
+VALIDATE_CREDENTIAL = {
+    'model_uid': 'fake-model-uid',
+    'server_url': 'http://127.0.0.1:9997/'
+}
+
+
+def encrypt_side_effect(tenant_id, encrypt_key):
+    return f'encrypted_{encrypt_key}'
+
+
+def decrypt_side_effect(tenant_id, encrypted_key):
+    return encrypted_key.replace('encrypted_', '')
+
+
+def test_is_credentials_valid_or_raise_valid(mocker):
+    mocker.patch('core.third_party.langchain.llms.xinference_llm.XinferenceLLM._call',
+                 return_value="abc")
+
+    MODEL_PROVIDER_CLASS.is_model_credentials_valid_or_raise(
+        model_name='username/test_model_name',
+        model_type=ModelType.TEXT_GENERATION,
+        credentials=VALIDATE_CREDENTIAL.copy()
+    )
+
+
+def test_is_credentials_valid_or_raise_invalid():
+    # raise CredentialsValidateFailedError if replicate_api_token is not in credentials
+    with pytest.raises(CredentialsValidateFailedError):
+        MODEL_PROVIDER_CLASS.is_model_credentials_valid_or_raise(
+            model_name='test_model_name',
+            model_type=ModelType.TEXT_GENERATION,
+            credentials={}
+        )
+
+    # raise CredentialsValidateFailedError if replicate_api_token is invalid
+    with pytest.raises(CredentialsValidateFailedError):
+        MODEL_PROVIDER_CLASS.is_model_credentials_valid_or_raise(
+            model_name='test_model_name',
+            model_type=ModelType.TEXT_GENERATION,
+            credentials={'server_url': 'invalid'})
+
+
+@patch('core.helper.encrypter.encrypt_token', side_effect=encrypt_side_effect)
+def test_encrypt_model_credentials(mock_encrypt, mocker):
+    api_key = 'http://127.0.0.1:9997/'
+
+    mocker.patch('core.model_providers.providers.xinference_provider.XinferenceProvider._get_extra_credentials',
+                 return_value={
+                     'model_handle_type': 'generate',
+                     'model_format': 'ggmlv3'
+                 })
+
+    result = MODEL_PROVIDER_CLASS.encrypt_model_credentials(
+        tenant_id='tenant_id',
+        model_name='test_model_name',
+        model_type=ModelType.TEXT_GENERATION,
+        credentials=VALIDATE_CREDENTIAL.copy()
+    )
+    mock_encrypt.assert_called_with('tenant_id', api_key)
+    assert result['server_url'] == f'encrypted_{api_key}'
+
+
+@patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
+def test_get_model_credentials_custom(mock_decrypt, mocker):
+    provider = Provider(
+        id='provider_id',
+        tenant_id='tenant_id',
+        provider_name=PROVIDER_NAME,
+        provider_type=ProviderType.CUSTOM.value,
+        encrypted_config=None,
+        is_valid=True,
+    )
+
+    encrypted_credential = VALIDATE_CREDENTIAL.copy()
+    encrypted_credential['server_url'] = 'encrypted_' + encrypted_credential['server_url']
+
+    mock_query = MagicMock()
+    mock_query.filter.return_value.first.return_value = ProviderModel(
+        encrypted_config=json.dumps(encrypted_credential)
+    )
+    mocker.patch('extensions.ext_database.db.session.query', return_value=mock_query)
+
+    model_provider = MODEL_PROVIDER_CLASS(provider=provider)
+    result = model_provider.get_model_credentials(
+        model_name='test_model_name',
+        model_type=ModelType.TEXT_GENERATION
+    )
+    assert result['server_url'] == 'http://127.0.0.1:9997/'
+
+
+@patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
+def test_get_model_credentials_obfuscated(mock_decrypt, mocker):
+    provider = Provider(
+        id='provider_id',
+        tenant_id='tenant_id',
+        provider_name=PROVIDER_NAME,
+        provider_type=ProviderType.CUSTOM.value,
+        encrypted_config=None,
+        is_valid=True,
+    )
+
+    encrypted_credential = VALIDATE_CREDENTIAL.copy()
+    encrypted_credential['server_url'] = 'encrypted_' + encrypted_credential['server_url']
+
+    mock_query = MagicMock()
+    mock_query.filter.return_value.first.return_value = ProviderModel(
+        encrypted_config=json.dumps(encrypted_credential)
+    )
+    mocker.patch('extensions.ext_database.db.session.query', return_value=mock_query)
+
+    model_provider = MODEL_PROVIDER_CLASS(provider=provider)
+    result = model_provider.get_model_credentials(
+        model_name='test_model_name',
+        model_type=ModelType.TEXT_GENERATION,
+        obfuscated=True
+    )
+    middle_token = result['server_url'][6:-2]
+    assert len(middle_token) == max(len(VALIDATE_CREDENTIAL['server_url']) - 8, 0)
+    assert all(char == '*' for char in middle_token)

docker/docker-compose.yaml

@@ -2,7 +2,7 @@ version: '3.1'
 services:
   # API service
   api:
-    image: langgenius/dify-api:0.3.14
+    image: langgenius/dify-api:0.3.15
     restart: always
     environment:
       # Startup mode, 'api' starts the API server.
@@ -124,7 +124,7 @@ services:
   # worker service
   # The Celery worker for processing the queue.
   worker:
-    image: langgenius/dify-api:0.3.14
+    image: langgenius/dify-api:0.3.15
     restart: always
     environment:
       # Startup mode, 'worker' starts the Celery worker for processing the queue.
@@ -176,7 +176,7 @@ services:
 
   # Frontend web application.
   web:
-    image: langgenius/dify-web:0.3.14
+    image: langgenius/dify-web:0.3.15
     restart: always
     environment:
       EDITION: SELF_HOSTED

web/app/(commonLayout)/app/(appDetailLayout)/[appId]/overview/cardView.tsx

@@ -17,6 +17,7 @@ import type { App } from '@/types/app'
 import type { UpdateAppSiteCodeResponse } from '@/models/app'
 import { asyncRunSafe } from '@/utils'
 import { NEED_REFRESH_APP_LIST_KEY } from '@/config'
+import type { IAppCardProps } from '@/app/components/app/overview/appCard'
 
 export type ICardViewProps = {
   appId: string
@@ -68,7 +69,7 @@ const CardView: FC<ICardViewProps> = ({ appId }) => {
     handleError(err)
   }
 
-  const onSaveSiteConfig = async (params: any) => {
+  const onSaveSiteConfig: IAppCardProps['onSaveSiteConfig'] = async (params) => {
     const [err] = await asyncRunSafe<App>(
       updateAppSiteConfig({
         url: `/apps/${appId}/site`,

web/app/(commonLayout)/app/(appDetailLayout)/[appId]/overview/page.tsx

@@ -16,7 +16,6 @@ const Overview = async ({
   const { t } = await useTranslation(locale, 'app-overview')
   return (
     <div className="h-full px-16 py-6 overflow-scroll">
-      {/* <WelcomeBanner /> */}
       <ApikeyInfoPanel />
       <div className='flex flex-row items-center justify-between mb-4 text-xl text-gray-900'>
         {t('overview.title')}

web/app/(commonLayout)/app/(appDetailLayout)/[appId]/overview/welcome-banner.tsx

@@ -1,200 +0,0 @@
-'use client'
-import type { FC } from 'react'
-import React, { useState } from 'react'
-import { useContext } from 'use-context-selector'
-import { useTranslation } from 'react-i18next'
-import Link from 'next/link'
-import useSWR, { useSWRConfig } from 'swr'
-import { ArrowTopRightOnSquareIcon } from '@heroicons/react/24/outline'
-import { ExclamationCircleIcon } from '@heroicons/react/24/solid'
-import { debounce } from 'lodash-es'
-import Popover from '@/app/components/base/popover'
-import Button from '@/app/components/base/button'
-import Tag from '@/app/components/base/tag'
-import { ToastContext } from '@/app/components/base/toast'
-import { updateOpenAIKey, validateOpenAIKey } from '@/service/apps'
-import { fetchTenantInfo } from '@/service/common'
-import I18n from '@/context/i18n'
-
-type IStatusType = 'normal' | 'verified' | 'error' | 'error-api-key-exceed-bill'
-
-const STATUS_COLOR_MAP = {
-  'normal': { color: '', bgColor: 'bg-primary-50', borderColor: 'border-primary-100' },
-  'error': { color: 'text-red-600', bgColor: 'bg-red-50', borderColor: 'border-red-100' },
-  'verified': { color: '', bgColor: 'bg-green-50', borderColor: 'border-green-100' },
-  'error-api-key-exceed-bill': { color: 'text-red-600', bgColor: 'bg-red-50', borderColor: 'border-red-100' },
-}
-
-const CheckCircleIcon: FC<{ className?: string }> = ({ className }) => {
-  return <svg width="20" height="20" viewBox="0 0 20 20" fill="none" xmlns="http://www.w3.org/2000/svg" className={className ?? ''}>
-    <rect width="20" height="20" rx="10" fill="#DEF7EC" />
-    <path fillRule="evenodd" clipRule="evenodd" d="M14.6947 6.70495C14.8259 6.83622 14.8996 7.01424 14.8996 7.19985C14.8996 7.38547 14.8259 7.56348 14.6947 7.69475L9.0947 13.2948C8.96343 13.426 8.78541 13.4997 8.5998 13.4997C8.41418 13.4997 8.23617 13.426 8.1049 13.2948L5.3049 10.4948C5.17739 10.3627 5.10683 10.1859 5.10842 10.0024C5.11002 9.81883 5.18364 9.64326 5.31342 9.51348C5.44321 9.38369 5.61878 9.31007 5.80232 9.30848C5.98585 9.30688 6.16268 9.37744 6.2947 9.50495L8.5998 11.8101L13.7049 6.70495C13.8362 6.57372 14.0142 6.5 14.1998 6.5C14.3854 6.5 14.5634 6.57372 14.6947 6.70495Z" fill="#046C4E" />
-  </svg>
-}
-
-type IEditKeyDiv = {
-  className?: string
-  showInPopover?: boolean
-  onClose?: () => void
-  getTenantInfo?: () => void
-}
-
-const EditKeyDiv: FC<IEditKeyDiv> = ({ className = '', showInPopover = false, onClose, getTenantInfo }) => {
-  const [inputValue, setInputValue] = useState<string | undefined>()
-  const [editStatus, setEditStatus] = useState<IStatusType>('normal')
-  const [loading, setLoading] = useState(false)
-  const [validating, setValidating] = useState(false)
-  const { notify } = useContext(ToastContext)
-  const { t } = useTranslation()
-  const { locale } = useContext(I18n)
-
-  // Hide the pop-up window and need to get the latest key again
-  // If the key is valid, the edit button will be hidden later
-  const onClosePanel = () => {
-    getTenantInfo && getTenantInfo()
-    onClose && onClose()
-  }
-
-  const onSaveKey = async () => {
-    if (editStatus === 'verified') {
-      setLoading(true)
-      try {
-        await updateOpenAIKey({ url: '/providers/openai/token', body: { token: inputValue ?? '' } })
-        notify({ type: 'success', message: t('common.actionMsg.modifiedSuccessfully') })
-        onClosePanel()
-      }
-      catch (err) {
-        notify({ type: 'error', message: t('common.actionMsg.modificationFailed') })
-      }
-      finally {
-        setLoading(false)
-      }
-    }
-  }
-
-  const validateKey = async (value: string) => {
-    try {
-      setValidating(true)
-      const res = await validateOpenAIKey({ url: '/providers/openai/token-validate', body: { token: value ?? '' } })
-      setEditStatus(res.result === 'success' ? 'verified' : 'error')
-    }
-    catch (err: any) {
-      if (err.status === 400) {
-        err.json().then(({ code }: any) => {
-          if (code === 'provider_request_failed')
-            setEditStatus('error-api-key-exceed-bill')
-        })
-      }
-      else {
-        setEditStatus('error')
-      }
-    }
-    finally {
-      setValidating(false)
-    }
-  }
-  const renderErrorMessage = () => {
-    if (validating) {
-      return (
-        <div className={'text-primary-600 mt-2 text-xs'}>
-          {t('common.provider.validating')}
-        </div>
-      )
-    }
-    if (editStatus === 'error-api-key-exceed-bill') {
-      return (
-        <div className={'text-[#D92D20] mt-2 text-xs'}>
-          {t('common.provider.apiKeyExceedBill')}
-          {locale === 'en' ? ' ' : ''}
-          <Link
-            className='underline'
-            href="https://platform.openai.com/account/api-keys"
-            target={'_blank'}>
-            {locale === 'en' ? 'this link' : '这篇文档'}
-          </Link>
-        </div>
-      )
-    }
-    if (editStatus === 'error') {
-      return (
-        <div className={'text-[#D92D20] mt-2 text-xs'}>
-          {t('common.provider.invalidKey')}
-        </div>
-      )
-    }
-    return null
-  }
-
-  return (
-    <div className={`flex flex-col w-full rounded-lg px-8 py-6 border-solid border-[0.5px] ${className} ${Object.values(STATUS_COLOR_MAP[editStatus]).join(' ')}`}>
-      {!showInPopover && <p className='text-xl font-medium text-gray-800'>{t('appOverview.welcome.firstStepTip')}</p>}
-      <p className={`${showInPopover ? 'text-sm' : 'text-xl'} font-medium text-gray-800`}>{t('appOverview.welcome.enterKeyTip')} {showInPopover ? '' : '👇'}</p>
-      <div className='relative mt-2'>
-        <input type="text"
-          className={`h-9 w-96 max-w-full py-2 pl-2 text-gray-900 rounded-lg bg-white sm:text-xs focus:ring-blue-500 focus:border-blue-500 shadow-sm ${editStatus === 'normal' ? 'pr-2' : 'pr-8'}`}
-          placeholder={t('appOverview.welcome.placeholder') || ''}
-          onChange={debounce((e) => {
-            setInputValue(e.target.value)
-            if (!e.target.value) {
-              setEditStatus('normal')
-              return
-            }
-            validateKey(e.target.value)
-          }, 300)}
-        />
-        {editStatus === 'verified' && <div className="absolute inset-y-0 right-0 flex flex-row-reverse items-center pr-6 pointer-events-none">
-          <CheckCircleIcon className="rounded-lg" />
-        </div>}
-        {(editStatus === 'error' || editStatus === 'error-api-key-exceed-bill') && <div className="absolute inset-y-0 right-0 flex flex-row-reverse items-center pr-6 pointer-events-none">
-          <ExclamationCircleIcon className="w-5 h-5 text-red-800" />
-        </div>}
-        {showInPopover ? null : <Button type='primary' onClick={onSaveKey} className='!h-9 !inline-block ml-2' loading={loading} disabled={editStatus !== 'verified'}>{t('common.operation.save')}</Button>}
-      </div>
-      {renderErrorMessage()}
-      <Link className="inline-flex items-center mt-2 text-xs font-normal cursor-pointer text-primary-600 w-fit" href="https://platform.openai.com/account/api-keys" target={'_blank'}>
-        {t('appOverview.welcome.getKeyTip')}
-        <ArrowTopRightOnSquareIcon className='w-3 h-3 ml-1 text-primary-600' aria-hidden="true" />
-      </Link>
-      {showInPopover && <div className='flex justify-end mt-6'>
-        <Button className='flex-shrink-0 mr-2' onClick={onClosePanel}>{t('common.operation.cancel')}</Button>
-        <Button type='primary' className='flex-shrink-0' onClick={onSaveKey} loading={loading} disabled={editStatus !== 'verified'}>{t('common.operation.save')}</Button>
-      </div>}
-    </div>
-  )
-}
-
-const WelcomeBanner: FC = () => {
-  const { data: userInfo } = useSWR({ url: '/info' }, fetchTenantInfo)
-  if (!userInfo)
-    return null
-  return userInfo?.providers?.find(({ token_is_set }) => token_is_set) ? null : <EditKeyDiv className='mb-8' />
-}
-
-export const EditKeyPopover: FC = () => {
-  const { data: userInfo } = useSWR({ url: '/info' }, fetchTenantInfo)
-  const { mutate } = useSWRConfig()
-  if (!userInfo)
-    return null
-
-  const getTenantInfo = () => {
-    mutate({ url: '/info' })
-  }
-  // In this case, the edit button is displayed
-  const targetProvider = userInfo?.providers?.some(({ token_is_set, is_valid }) => token_is_set && is_valid)
-  return (
-    !targetProvider
-      ? <div className='flex items-center'>
-        <Tag className='mr-2 h-fit' color='red'><ExclamationCircleIcon className='h-3.5 w-3.5 mr-2' />OpenAI API key invalid</Tag>
-        <Popover
-          htmlContent={<EditKeyDiv className='!border-0' showInPopover={true} getTenantInfo={getTenantInfo} />}
-          trigger='click'
-          position='br'
-          btnElement='Edit'
-          btnClassName='text-primary-600 !text-xs px-3 py-1.5'
-          className='!p-0 !w-[464px] h-[200px]'
-        />
-      </div>
-      : null)
-}
-
-export default WelcomeBanner

web/app/(commonLayout)/apps/Apps.tsx

@@ -68,11 +68,11 @@ const Apps = () => {
 
   return (
     <nav className='grid content-start grid-cols-1 gap-4 px-12 pt-8 sm:grid-cols-2 md:grid-cols-3 lg:grid-cols-4 grow shrink-0'>
+      { isCurrentWorkspaceManager
+      && <NewAppCard ref={anchorRef} onSuccess={mutate} />}
       {data?.map(({ data: apps }) => apps.map(app => (
         <AppCard key={app.id} app={app} onRefresh={mutate} />
       )))}
-      { isCurrentWorkspaceManager
-      && <NewAppCard ref={anchorRef} onSuccess={mutate} />}
       {
         showPayStatusModal && (
           <Confirm

web/app/(commonLayout)/datasets/Datasets.tsx

@@ -42,10 +42,10 @@ const Datasets = () => {
 
   return (
     <nav className='grid content-start grid-cols-1 gap-4 px-12 pt-8 sm:grid-cols-2 md:grid-cols-3 lg:grid-cols-4 grow shrink-0'>
+      { isCurrentWorkspaceManager && <NewDatasetCard ref={anchorRef} /> }
       {data?.map(({ data: datasets }) => datasets.map(dataset => (
         <DatasetCard key={dataset.id} dataset={dataset} onDelete={mutate} />),
       ))}
-      { isCurrentWorkspaceManager && <NewDatasetCard ref={anchorRef} /> }
     </nav>
   )
 }

web/app/activate/activateForm.tsx

@@ -1,5 +1,5 @@
 'use client'
-import { useState } from 'react'
+import { useCallback, useState } from 'react'
 import { useContext } from 'use-context-selector'
 import { useTranslation } from 'react-i18next'
 import useSWR from 'swr'
@@ -22,7 +22,7 @@ const validPassword = /^(?=.*[a-zA-Z])(?=.*\d).{8,}$/
 
 const ActivateForm = () => {
   const { t } = useTranslation()
-  const { locale } = useContext(I18n)
+  const { locale, setLocaleOnClient } = useContext(I18n)
   const searchParams = useSearchParams()
   const workspaceID = searchParams.get('workspace_id')
   const email = searchParams.get('email')
@@ -45,14 +45,16 @@ const ActivateForm = () => {
   const [timezone, setTimezone] = useState('Asia/Shanghai')
   const [language, setLanguage] = useState('en-US')
   const [showSuccess, setShowSuccess] = useState(false)
+  const defaultLanguage = useCallback(() => (window.navigator.language.startsWith('zh') ? languageMaps['zh-Hans'] : languageMaps.en) || languageMaps.en, [])
 
-  const showErrorMessage = (message: string) => {
+  const showErrorMessage = useCallback((message: string) => {
     Toast.notify({
       type: 'error',
       message,
     })
-  }
-  const valid = () => {
+  }, [])
+
+  const valid = useCallback(() => {
     if (!name.trim()) {
       showErrorMessage(t('login.error.nameEmpty'))
       return false
@@ -65,9 +67,9 @@ const ActivateForm = () => {
       showErrorMessage(t('login.error.passwordInvalid'))
 
     return true
-  }
+  }, [name, password, showErrorMessage, t])
 
-  const handleActivate = async () => {
+  const handleActivate = useCallback(async () => {
     if (!valid())
       return
     try {
@@ -83,17 +85,18 @@ const ActivateForm = () => {
           timezone,
         },
       })
+      setLocaleOnClient(language.startsWith('en') ? 'en' : 'zh-Hans', false)
       setShowSuccess(true)
     }
     catch {
       recheck()
     }
-  }
+  }, [email, language, name, password, recheck, setLocaleOnClient, timezone, token, valid, workspaceID])
 
   return (
     <div className={
       cn(
-        'flex flex-col items-center w-full grow items-center justify-center',
+        'flex flex-col items-center w-full grow justify-center',
         'px-6',
         'md:px-[108px]',
       )
@@ -167,7 +170,7 @@ const ActivateForm = () => {
                 </label>
                 <div className="relative mt-1 rounded-md shadow-sm">
                   <SimpleSelect
-                    defaultValue={languageMaps.en}
+                    defaultValue={defaultLanguage()}
                     items={languages}
                     onSelect={(item) => {
                       setLanguage(item.value as string)

web/app/components/app/chat/answer/index.tsx

@@ -225,7 +225,7 @@ const Answer: FC<IAnswerProps> = ({ item, feedbackDisabled = false, isHideFeedba
                           setLoading(true)
                           const res = await onSubmitAnnotation?.(id, inputValue)
                           if (res)
-                            setAnnotation({ ...annotation, content: inputValue } as any)
+                            setAnnotation({ ...annotation, content: inputValue } as Annotation)
                           setLoading(false)
                           setShowEdit(false)
                         }}>{t('common.operation.confirm')}</Button>

web/app/components/app/chat/index.tsx

@@ -81,7 +81,7 @@ const Chat: FC<IChatProps> = ({
   const isUseInputMethod = useRef(false)
 
   const [query, setQuery] = React.useState('')
-  const handleContentChange = (e: any) => {
+  const handleContentChange = (e: React.ChangeEvent<HTMLTextAreaElement>) => {
     const value = e.target.value
     setQuery(value)
   }
@@ -111,7 +111,7 @@ const Chat: FC<IChatProps> = ({
       setQuery('')
   }
 
-  const handleKeyUp = (e: any) => {
+  const handleKeyUp = (e: React.KeyboardEvent<HTMLTextAreaElement>) => {
     if (e.code === 'Enter') {
       e.preventDefault()
       // prevent send message when using input method enter
@@ -120,7 +120,7 @@ const Chat: FC<IChatProps> = ({
     }
   }
 
-  const handleKeyDown = (e: any) => {
+  const handleKeyDown = (e: React.KeyboardEvent<HTMLTextAreaElement>) => {
     isUseInputMethod.current = e.nativeEvent.isComposing
     if (e.code === 'Enter' && !e.shiftKey) {
       setQuery(query.replace(/\n$/, ''))

web/app/components/app/configuration/config-var/input-type-icon.tsx

@@ -1,19 +1,19 @@
 'use client'
-import React, { FC, ReactNode } from 'react'
-import { ReactElement } from 'react-markdown/lib/react-markdown'
+import React from 'react'
+import type { FC } from 'react'
 
-export interface IInputTypeIconProps {
-  type: string
+type IInputTypeIconProps = {
+  type: 'string' | 'select'
 }
 
-const IconMap = (type: string) => {
-  const icons: Record<string, ReactNode> = {
-    'string': (
+const IconMap = (type: IInputTypeIconProps['type']) => {
+  const icons = {
+    string: (
       <svg width="12" height="12" viewBox="0 0 12 12" fill="none" xmlns="http://www.w3.org/2000/svg">
         <path fillRule="evenodd" clipRule="evenodd" d="M3.52593 0.166672H8.47411C8.94367 0.166665 9.33123 0.166659 9.64692 0.192452C9.97481 0.219242 10.2762 0.276738 10.5593 0.420991C10.9984 0.644695 11.3553 1.00165 11.579 1.44069C11.7233 1.72381 11.7808 2.02522 11.8076 2.35311C11.8334 2.6688 11.8334 3.05634 11.8334 3.5259V8.47411C11.8334 8.94367 11.8334 9.33121 11.8076 9.6469C11.7808 9.97479 11.7233 10.2762 11.579 10.5593C11.3553 10.9984 10.9984 11.3553 10.5593 11.579C10.2762 11.7233 9.97481 11.7808 9.64692 11.8076C9.33123 11.8334 8.94369 11.8333 8.47413 11.8333H3.52592C3.05636 11.8333 2.66882 11.8334 2.35312 11.8076C2.02523 11.7808 1.72382 11.7233 1.44071 11.579C1.00167 11.3553 0.644711 10.9984 0.421006 10.5593C0.276753 10.2762 0.219257 9.97479 0.192468 9.6469C0.166674 9.33121 0.16668 8.94366 0.166687 8.4741V3.52591C0.16668 3.05635 0.166674 2.6688 0.192468 2.35311C0.219257 2.02522 0.276753 1.72381 0.421006 1.44069C0.644711 1.00165 1.00167 0.644695 1.44071 0.420991C1.72382 0.276738 2.02523 0.219242 2.35312 0.192452C2.66882 0.166659 3.05637 0.166665 3.52593 0.166672ZM3.08335 3.08334C3.08335 2.76117 3.34452 2.50001 3.66669 2.50001H8.33335C8.65552 2.50001 8.91669 2.76117 8.91669 3.08334C8.91669 3.4055 8.65552 3.66667 8.33335 3.66667H6.58335V8.91667C6.58335 9.23884 6.32219 9.5 6.00002 9.5C5.67785 9.5 5.41669 9.23884 5.41669 8.91667V3.66667H3.66669C3.34452 3.66667 3.08335 3.4055 3.08335 3.08334Z" fill="#98A2B3" />
       </svg>
     ),
-    'select': (
+    select: (
       <svg width="14" height="14" viewBox="0 0 14 14" fill="none" xmlns="http://www.w3.org/2000/svg">
         <path fillRule="evenodd" clipRule="evenodd" d="M7.48913 4.08334H3.01083C2.70334 4.08333 2.43804 4.08333 2.21955 4.10118C1.98893 4.12002 1.75955 4.16162 1.53883 4.27408C1.20955 4.44186 0.941831 4.70958 0.774053 5.03886C0.66159 5.25958 0.619989 5.48896 0.601147 5.71958C0.583295 5.93807 0.583304 6.20334 0.583313 6.51084V10.9892C0.583304 11.2967 0.583295 11.5619 0.601147 11.7804C0.619989 12.0111 0.66159 12.2404 0.774053 12.4612C0.941831 12.7904 1.20955 13.0582 1.53883 13.2259C1.75955 13.3384 1.98893 13.38 2.21955 13.3988C2.43803 13.4167 2.70329 13.4167 3.01077 13.4167H7.48912C7.7966 13.4167 8.06193 13.4167 8.28041 13.3988C8.51103 13.38 8.74041 13.3384 8.96113 13.2259C9.29041 13.0582 9.55813 12.7904 9.72591 12.4612C9.83837 12.2404 9.87997 12.0111 9.89882 11.7804C9.91667 11.5619 9.91666 11.2967 9.91665 10.9892V6.51087C9.91666 6.20336 9.91667 5.93808 9.89882 5.71958C9.87997 5.48896 9.83837 5.25958 9.72591 5.03886C9.55813 4.70958 9.29041 4.44186 8.96113 4.27408C8.74041 4.16162 8.51103 4.12002 8.28041 4.10118C8.06192 4.08333 7.79663 4.08333 7.48913 4.08334ZM7.70413 7.70416C7.93193 7.47635 7.93193 7.107 7.70413 6.8792C7.47632 6.65139 7.10697 6.65139 6.87917 6.8792L4.66665 9.09172L3.91246 8.33753C3.68465 8.10973 3.31531 8.10973 3.0875 8.33753C2.8597 8.56534 2.8597 8.93468 3.0875 9.16249L4.25417 10.3292C4.48197 10.557 4.85132 10.557 5.07913 10.3292L7.70413 7.70416Z" fill="#98A2B3" />
         <path d="M10.9891 0.583344H6.51083C6.20334 0.583334 5.93804 0.583326 5.71955 0.601177C5.48893 0.620019 5.25955 0.66162 5.03883 0.774084C4.70955 0.941862 4.44183 1.20958 4.27405 1.53886C4.16159 1.75958 4.11999 1.98896 4.10115 2.21958C4.08514 2.41545 4.08349 2.64892 4.08333 2.91669L7.51382 2.91668C7.79886 2.91662 8.10791 2.91654 8.37541 2.9384C8.67818 2.96314 9.07818 3.02436 9.49078 3.23459C10.0396 3.51422 10.4858 3.96041 10.7654 4.50922C10.9756 4.92182 11.0369 5.32182 11.0616 5.62459C11.0835 5.8921 11.0834 6.20115 11.0833 6.48619L11.0833 9.91666C11.3511 9.9165 11.5845 9.91485 11.7804 9.89885C12.011 9.88 12.2404 9.8384 12.4611 9.72594C12.7904 9.55816 13.0581 9.29045 13.2259 8.96116C13.3384 8.74044 13.38 8.51106 13.3988 8.28044C13.4167 8.06196 13.4167 7.7967 13.4166 7.48922V3.01087C13.4167 2.70339 13.4167 2.43807 13.3988 2.21958C13.38 1.98896 13.3384 1.75958 13.2259 1.53886C13.0581 1.20958 12.7904 0.941862 12.4611 0.774084C12.2404 0.66162 12.011 0.620019 11.7804 0.601177C11.5619 0.583326 11.2966 0.583334 10.9891 0.583344Z" fill="#98A2B3" />
@@ -21,11 +21,11 @@ const IconMap = (type: string) => {
     ),
   }
 
-  return icons[type] as any
+  return icons[type]
 }
 
 const InputTypeIcon: FC<IInputTypeIconProps> = ({
-  type
+  type,
 }) => {
   const Icon = IconMap(type)
   return Icon

web/app/components/app/configuration/config/index.tsx

@@ -38,7 +38,7 @@ const Config: FC = () => {
     setSpeechToTextConfig,
   } = useContext(ConfigContext)
   const isChatApp = mode === AppType.chat
-  const { currentProvider } = useProviderContext()
+  const { speech2textDefaultModel } = useProviderContext()
 
   const promptTemplate = modelConfig.configs.prompt_template
   const promptVariables = modelConfig.configs.prompt_variables
@@ -90,7 +90,7 @@ const Config: FC = () => {
     },
   })
 
-  const hasChatConfig = isChatApp && (featureConfig.openingStatement || featureConfig.suggestedQuestionsAfterAnswer || (featureConfig.speechToText && currentProvider?.provider_name === 'openai'))
+  const hasChatConfig = isChatApp && (featureConfig.openingStatement || featureConfig.suggestedQuestionsAfterAnswer || (featureConfig.speechToText && !!speech2textDefaultModel))
   const hasToolbox = false
 
   const [showAutomatic, { setTrue: showAutomaticTrue, setFalse: showAutomaticFalse }] = useBoolean(false)
@@ -120,7 +120,7 @@ const Config: FC = () => {
             isChatApp={isChatApp}
             config={featureConfig}
             onChange={handleFeatureChange}
-            showSpeechToTextItem={currentProvider?.provider_name === 'openai'}
+            showSpeechToTextItem={!!speech2textDefaultModel}
           />
         )}
         {showAutomatic && (
@@ -160,7 +160,7 @@ const Config: FC = () => {
                 }
               }
               isShowSuggestedQuestionsAfterAnswer={featureConfig.suggestedQuestionsAfterAnswer}
-              isShowSpeechText={featureConfig.speechToText && currentProvider?.provider_name === 'openai'}
+              isShowSpeechText={featureConfig.speechToText && !!speech2textDefaultModel}
             />
           )
         }

web/app/components/app/configuration/dataset-config/card-item/index.tsx

@@ -6,22 +6,17 @@ import { useTranslation } from 'react-i18next'
 import TypeIcon from '../type-icon'
 import RemoveIcon from '../../base/icons/remove-icon'
 import s from './style.module.css'
+import type { DataSet } from '@/models/datasets'
 import { formatNumber } from '@/utils/format'
 import Tooltip from '@/app/components/base/tooltip'
 
 export type ICardItemProps = {
   className?: string
-  config: any
+  config: DataSet
   onRemove: (id: string) => void
   readonly?: boolean
 }
 
-// const RemoveIcon = ({ className, onClick }: { className: string, onClick: () => void }) => (
-//   <svg className={className} onClick={onClick} width="24" height="24" viewBox="0 0 24 24" fill="none" xmlns="http://www.w3.org/2000/svg">
-//     <path d="M10 6H14M6 8H18M16.6667 8L16.1991 15.0129C16.129 16.065 16.0939 16.5911 15.8667 16.99C15.6666 17.3412 15.3648 17.6235 15.0011 17.7998C14.588 18 14.0607 18 13.0062 18H10.9938C9.93927 18 9.41202 18 8.99889 17.7998C8.63517 17.6235 8.33339 17.3412 8.13332 16.99C7.90607 16.5911 7.871 16.065 7.80086 15.0129L7.33333 8M10.6667 11V14.3333M13.3333 11V14.3333" stroke="#667085" strokeWidth="1.5" strokeLinecap="round" strokeLinejoin="round" />
-//   </svg>
-// )
-
 const CardItem: FC<ICardItemProps> = ({
   className,
   config,

web/app/components/app/configuration/debug/index.tsx

@@ -52,7 +52,7 @@ const Debug: FC<IDebug> = ({
     modelConfig,
     completionParams,
   } = useContext(ConfigContext)
-  const { currentProvider } = useProviderContext()
+  const { speech2textDefaultModel } = useProviderContext()
   const [chatList, setChatList, getChatList] = useGetState<IChatItem[]>([])
   const chatListDomRef = useRef<HTMLDivElement>(null)
   useEffect(() => {
@@ -390,7 +390,7 @@ const Debug: FC<IDebug> = ({
                   }}
                   isShowSuggestion={doShowSuggestion}
                   suggestionList={suggestQuestions}
-                  isShowSpeechToText={speechToTextConfig.enabled && currentProvider?.provider_name === 'openai'}
+                  isShowSpeechToText={speechToTextConfig.enabled && !!speech2textDefaultModel}
                 />
               </div>
             </div>

web/app/components/app/configuration/index.tsx

@@ -16,22 +16,21 @@ import ConfigModel from '@/app/components/app/configuration/config-model'
 import Config from '@/app/components/app/configuration/config'
 import Debug from '@/app/components/app/configuration/debug'
 import Confirm from '@/app/components/base/confirm'
-import { ProviderType } from '@/types/app'
+import { ProviderEnum } from '@/app/components/header/account-setting/model-page/declarations'
 import type { AppDetailResponse } from '@/models/app'
 import { ToastContext } from '@/app/components/base/toast'
-import { fetchTenantInfo } from '@/service/common'
 import { fetchAppDetail, updateAppModelConfig } from '@/service/apps'
 import { promptVariablesToUserInputsForm, userInputsFormToPromptVariables } from '@/utils/model-config'
 import { fetchDatasets } from '@/service/datasets'
 import AccountSetting from '@/app/components/header/account-setting'
+import { useProviderContext } from '@/context/provider-context'
 
 const Configuration: FC = () => {
   const { t } = useTranslation()
   const { notify } = useContext(ToastContext)
 
   const [hasFetchedDetail, setHasFetchedDetail] = useState(false)
-  const [hasFetchedKey, setHasFetchedKey] = useState(false)
-  const isLoading = !hasFetchedDetail || !hasFetchedKey
+  const isLoading = !hasFetchedDetail
   const pathname = usePathname()
   const matched = pathname.match(/\/app\/([^/]+)/)
   const appId = (matched?.length && matched[1]) ? matched[1] : ''
@@ -68,7 +67,7 @@ const Configuration: FC = () => {
     frequency_penalty: 1, // -2-2
   })
   const [modelConfig, doSetModelConfig] = useState<ModelConfig>({
-    provider: ProviderType.openai,
+    provider: ProviderEnum.openai,
     model_id: 'gpt-3.5-turbo',
     configs: {
       prompt_template: '',
@@ -85,7 +84,7 @@ const Configuration: FC = () => {
     doSetModelConfig(newModelConfig)
   }
 
-  const setModelId = (modelId: string, provider: ProviderType) => {
+  const setModelId = (modelId: string, provider: ProviderEnum) => {
     const newModelConfig = produce(modelConfig, (draft: any) => {
       draft.provider = provider
       draft.model_id = modelId
@@ -113,25 +112,27 @@ const Configuration: FC = () => {
     })
   }
 
-  const [hasSetCustomAPIKEY, setHasSetCustomerAPIKEY] = useState(true)
-  const [isTrailFinished, setIsTrailFinished] = useState(false)
+  const { textGenerationModelList } = useProviderContext()
+  const hasSetCustomAPIKEY = !!textGenerationModelList?.find(({ model_provider: provider }) => {
+    if (provider.provider_type === 'system' && provider.quota_type === 'paid')
+      return true
+
+    if (provider.provider_type === 'custom')
+      return true
+
+    return false
+  })
+  const isTrailFinished = !hasSetCustomAPIKEY && textGenerationModelList
+    .filter(({ model_provider: provider }) => provider.quota_type === 'trial')
+    .every(({ model_provider: provider }) => {
+      const { quota_used, quota_limit } = provider
+      return quota_used === quota_limit
+    })
+
   const hasSetAPIKEY = hasSetCustomAPIKEY || !isTrailFinished
 
   const [isShowSetAPIKey, { setTrue: showSetAPIKey, setFalse: hideSetAPIkey }] = useBoolean()
 
-  const checkAPIKey = async () => {
-    const { in_trail, trial_end_reason } = await fetchTenantInfo({ url: '/info' })
-    const isTrailFinished = in_trail && trial_end_reason === 'trial_exceeded'
-    const hasSetCustomAPIKEY = trial_end_reason === 'using_custom'
-    setHasSetCustomerAPIKEY(hasSetCustomAPIKEY)
-    setIsTrailFinished(isTrailFinished)
-    setHasFetchedKey(true)
-  }
-
-  useEffect(() => {
-    checkAPIKey()
-  }, [])
-
   useEffect(() => {
     (fetchAppDetail({ url: '/apps', id: appId }) as any).then(async (res: AppDetailResponse) => {
       setMode(res.mode)
@@ -284,7 +285,7 @@ const Configuration: FC = () => {
               {/* Model and Parameters */}
               <ConfigModel
                 mode={mode}
-                provider={modelConfig.provider as ProviderType}
+                provider={modelConfig.provider as ProviderEnum}
                 completionParams={completionParams}
                 modelId={modelConfig.model_id}
                 setModelId={setModelId}
@@ -338,7 +339,6 @@ const Configuration: FC = () => {
           )
         }
         {isShowSetAPIKey && <AccountSetting activeTab="provider" onCancel={async () => {
-          await checkAPIKey()
           hideSetAPIkey()
         }} />}
       </>

web/app/components/app/overview/appCard.tsx

@@ -13,6 +13,7 @@ import SettingsModal from './settings'
 import EmbeddedModal from './embedded'
 import CustomizeModal from './customize'
 import style from './style.module.css'
+import type { ConfigParams } from './settings'
 import Tooltip from '@/app/components/base/tooltip'
 import AppBasic from '@/app/components/app-sidebar/basic'
 import { asyncRunSafe, randomString } from '@/utils'
@@ -31,9 +32,9 @@ export type IAppCardProps = {
   appInfo: AppDetailResponse
   cardType?: 'api' | 'webapp'
   customBgColor?: string
-  onChangeStatus: (val: boolean) => Promise<any>
-  onSaveSiteConfig?: (params: any) => Promise<any>
-  onGenerateCode?: () => Promise<any>
+  onChangeStatus: (val: boolean) => Promise<void>
+  onSaveSiteConfig?: (params: ConfigParams) => Promise<void>
+  onGenerateCode?: () => Promise<void>
 }
 
 const EmbedIcon: FC<{ className?: string }> = ({ className = '' }) => {
@@ -193,7 +194,7 @@ function AppCard({
         </div>
         <div className={'pt-2 flex flex-row items-center'}>
           {!isApp && <SecretKeyButton className='flex-shrink-0 !h-8 bg-white mr-2' textCls='!text-gray-700 font-medium' iconCls='stroke-[1.2px]' appId={appInfo.id} />}
-          {OPERATIONS_MAP[cardType].map((op: any) => {
+          {OPERATIONS_MAP[cardType].map((op) => {
             const disabled
               = op.opName === t('appOverview.overview.appInfo.settings.entry')
                 ? false

web/app/components/app/overview/settings/index.tsx

@@ -18,7 +18,7 @@ export type ISettingsModalProps = {
   isShow: boolean
   defaultValue?: string
   onClose: () => void
-  onSave?: (params: ConfigParams) => Promise<any>
+  onSave?: (params: ConfigParams) => Promise<void>
 }
 
 export type ConfigParams = {
@@ -26,6 +26,10 @@ export type ConfigParams = {
   description: string
   default_language: string
   prompt_public: boolean
+  copyright: string
+  privacy_policy: string
+  icon: string
+  icon_background: string
 }
 
 const LANGUAGE_MAP: Record<Language, string> = {
@@ -82,7 +86,7 @@ const SettingsModal: FC<ISettingsModalProps> = ({
   }
 
   const onChange = (field: string) => {
-    return (e: any) => {
+    return (e: React.ChangeEvent<HTMLInputElement | HTMLTextAreaElement>) => {
       setInputInfo(item => ({ ...item, [field]: e.target.value }))
     }
   }

web/app/components/app/text-generate/item/index.tsx

@@ -10,7 +10,7 @@ import { HashtagIcon } from '@heroicons/react/24/solid'
 import { Markdown } from '@/app/components/base/markdown'
 import Loading from '@/app/components/base/loading'
 import Toast from '@/app/components/base/toast'
-import type { Feedbacktype } from '@/app/components/app/chat'
+import type { Feedbacktype } from '@/app/components/app/chat/type'
 import { fetchMoreLikeThis, updateFeedback } from '@/service/share'
 
 const MAX_DEPTH = 3
@@ -136,7 +136,7 @@ const GenerationItem: FC<IGenerationItemProps> = ({
   }
 
   const mainStyle = (() => {
-    const res: any = !isTop
+    const res: React.CSSProperties = !isTop
       ? {
         background: depth % 2 === 0 ? 'linear-gradient(90.07deg, #F9FAFB 0.05%, rgba(249, 250, 251, 0) 99.93%)' : '#fff',
       }

web/app/components/base/button/index.tsx

@@ -7,6 +7,7 @@ export type IButtonProps = {
   className?: string
   disabled?: boolean
   loading?: boolean
+  tabIndex?: number
   children: React.ReactNode
   onClick?: MouseEventHandler<HTMLDivElement>
 }
@@ -18,6 +19,7 @@ const Button: FC<IButtonProps> = ({
   className,
   onClick,
   loading = false,
+  tabIndex,
 }) => {
   let style = 'cursor-pointer'
   switch (type) {
@@ -35,6 +37,7 @@ const Button: FC<IButtonProps> = ({
   return (
     <div
       className={`inline-flex justify-center items-center content-center h-9 leading-5 rounded-lg px-4 py-2 text-base ${style} ${className && className}`}
+      tabIndex={tabIndex}
       onClick={disabled ? undefined : onClick}
     >
       {children}

web/app/components/base/icons/assets/public/llm/xorbits-inference-text.svg

@@ -0,0 +1,42 @@
+<svg width="152" height="24" viewBox="0 0 152 24" fill="none" xmlns="http://www.w3.org/2000/svg">
+<g id="xorbits 1" clip-path="url(#clip0_9866_6170)">
+<path id="Vector" d="M8.00391 12.3124C8.69334 13.0754 9.47526 13.7494 10.3316 14.3188C11.0667 14.8105 11.8509 15.2245 12.6716 15.5541C14.1617 14.1465 15.3959 12.4907 16.3192 10.6606L21.7051 0L12.3133 7.38353C10.5832 8.74456 9.12178 10.416 8.00391 12.3124Z" fill="url(#paint0_linear_9866_6170)"/>
+<path id="Vector_2" d="M7.23504 18.9512C6.56092 18.5012 5.92386 18.0265 5.3221 17.5394L2.06445 24L7.91975 19.3959C7.69034 19.2494 7.46092 19.103 7.23504 18.9512Z" fill="url(#paint1_linear_9866_6170)"/>
+<path id="Vector_3" d="M19.3161 8.57474C21.0808 10.9147 21.5961 13.5159 20.3996 15.3053C18.6526 17.9189 13.9161 17.8183 9.82024 15.0812C5.72435 12.3441 3.82024 8.0065 5.56729 5.39297C6.76377 3.60356 9.36318 3.0865 12.2008 3.81886C7.29318 1.73474 2.62376 1.94121 0.813177 4.64474C-1.45976 8.04709 1.64435 14.1177 7.74494 18.1889C13.8455 22.26 20.6361 22.8124 22.9091 19.4118C24.7179 16.703 23.1173 12.3106 19.3161 8.57474Z" fill="url(#paint2_linear_9866_6170)"/>
+<g id="Xorbits Inference">
+<path d="M35.5162 12.142L38.5402 17H36.7482L34.5502 13.472L32.4922 17H30.7142L33.7382 12.142L30.7002 7.27002H32.4922L34.7042 10.826L36.7762 7.27002H38.5542L35.5162 12.142Z" fill="#1D2939"/>
+<path d="M43.3584 17.126C42.6304 17.126 41.9724 16.9627 41.3844 16.636C40.7964 16.3 40.3344 15.8334 39.9984 15.236C39.6624 14.6294 39.4944 13.9293 39.4944 13.136C39.4944 12.352 39.6671 11.6567 40.0124 11.05C40.3577 10.4434 40.8291 9.97668 41.4264 9.65002C42.0237 9.32335 42.6911 9.16002 43.4284 9.16002C44.1657 9.16002 44.8331 9.32335 45.4304 9.65002C46.0277 9.97668 46.4991 10.4434 46.8444 11.05C47.1897 11.6567 47.3624 12.352 47.3624 13.136C47.3624 13.92 47.185 14.6154 46.8304 15.222C46.4757 15.8287 45.9904 16.3 45.3744 16.636C44.7677 16.9627 44.0957 17.126 43.3584 17.126ZM43.3584 15.74C43.769 15.74 44.1517 15.642 44.5064 15.446C44.8704 15.25 45.1644 14.956 45.3884 14.564C45.6124 14.172 45.7244 13.696 45.7244 13.136C45.7244 12.576 45.6171 12.1047 45.4024 11.722C45.1877 11.33 44.9031 11.036 44.5484 10.84C44.1937 10.644 43.8111 10.546 43.4004 10.546C42.9897 10.546 42.607 10.644 42.2524 10.84C41.9071 11.036 41.6317 11.33 41.4264 11.722C41.221 12.1047 41.1184 12.576 41.1184 13.136C41.1184 13.9667 41.3284 14.6107 41.7484 15.068C42.1777 15.516 42.7144 15.74 43.3584 15.74Z" fill="#1D2939"/>
+<path d="M50.2561 10.406C50.4895 10.014 50.7974 9.71068 51.1801 9.49602C51.5721 9.27202 52.0341 9.16002 52.5661 9.16002V10.812H52.1601C51.5348 10.812 51.0588 10.9707 50.7321 11.288C50.4148 11.6054 50.2561 12.156 50.2561 12.94V17H48.6601V9.28602H50.2561V10.406Z" fill="#1D2939"/>
+<path d="M55.3492 10.434C55.6198 10.0607 55.9885 9.75735 56.4552 9.52402C56.9312 9.28135 57.4585 9.16002 58.0372 9.16002C58.7185 9.16002 59.3345 9.32335 59.8852 9.65002C60.4358 9.97668 60.8698 10.4434 61.1872 11.05C61.5045 11.6473 61.6632 12.3333 61.6632 13.108C61.6632 13.8827 61.5045 14.578 61.1872 15.194C60.8698 15.8007 60.4312 16.2767 59.8712 16.622C59.3205 16.958 58.7092 17.126 58.0372 17.126C57.4398 17.126 56.9078 17.0093 56.4412 16.776C55.9838 16.5427 55.6198 16.244 55.3492 15.88V17H53.7532V6.64002H55.3492V10.434ZM60.0392 13.108C60.0392 12.576 59.9272 12.1187 59.7032 11.736C59.4885 11.344 59.1992 11.05 58.8352 10.854C58.4805 10.6487 58.0978 10.546 57.6872 10.546C57.2858 10.546 56.9032 10.6487 56.5392 10.854C56.1845 11.0594 55.8952 11.358 55.6712 11.75C55.4565 12.142 55.3492 12.604 55.3492 13.136C55.3492 13.668 55.4565 14.1347 55.6712 14.536C55.8952 14.928 56.1845 15.2267 56.5392 15.432C56.9032 15.6374 57.2858 15.74 57.6872 15.74C58.0978 15.74 58.4805 15.6374 58.8352 15.432C59.1992 15.2174 59.4885 14.9093 59.7032 14.508C59.9272 14.1067 60.0392 13.64 60.0392 13.108Z" fill="#1D2939"/>
+<path d="M63.7734 8.26402C63.4841 8.26402 63.2414 8.16602 63.0454 7.97002C62.8494 7.77402 62.7514 7.53135 62.7514 7.24202C62.7514 6.95268 62.8494 6.71002 63.0454 6.51402C63.2414 6.31802 63.4841 6.22002 63.7734 6.22002C64.0534 6.22002 64.2914 6.31802 64.4874 6.51402C64.6834 6.71002 64.7814 6.95268 64.7814 7.24202C64.7814 7.53135 64.6834 7.77402 64.4874 7.97002C64.2914 8.16602 64.0534 8.26402 63.7734 8.26402ZM64.5574 9.28602V17H62.9614V9.28602H64.5574Z" fill="#1D2939"/>
+<path d="M68.2348 10.588V14.858C68.2348 15.1474 68.3002 15.3573 68.4309 15.488C68.5709 15.6093 68.8042 15.67 69.1308 15.67H70.1109V17H68.8508C68.1322 17 67.5815 16.832 67.1988 16.496C66.8162 16.16 66.6248 15.614 66.6248 14.858V10.588H65.7148V9.28602H66.6248V7.36802H68.2348V9.28602H70.1109V10.588H68.2348Z" fill="#1D2939"/>
+<path d="M74.1018 17.126C73.4952 17.126 72.9492 17.0187 72.4638 16.804C71.9878 16.58 71.6098 16.2813 71.3298 15.908C71.0498 15.5253 70.9005 15.1007 70.8818 14.634H72.5338C72.5618 14.9607 72.7158 15.236 72.9958 15.46C73.2852 15.6747 73.6445 15.782 74.0738 15.782C74.5218 15.782 74.8672 15.698 75.1098 15.53C75.3618 15.3527 75.4878 15.1287 75.4878 14.858C75.4878 14.5687 75.3478 14.354 75.0678 14.214C74.7972 14.074 74.3632 13.92 73.7658 13.752C73.1872 13.5933 72.7158 13.4394 72.3518 13.29C71.9878 13.1407 71.6705 12.912 71.3998 12.604C71.1385 12.296 71.0078 11.89 71.0078 11.386C71.0078 10.9753 71.1292 10.602 71.3718 10.266C71.6145 9.92068 71.9598 9.65002 72.4078 9.45402C72.8652 9.25802 73.3878 9.16002 73.9758 9.16002C74.8532 9.16002 75.5578 9.38402 76.0898 9.83202C76.6312 10.2707 76.9205 10.8727 76.9578 11.638H75.3618C75.3338 11.2927 75.1938 11.0173 74.9418 10.812C74.6898 10.6067 74.3492 10.504 73.9198 10.504C73.4998 10.504 73.1778 10.5833 72.9538 10.742C72.7298 10.9007 72.6178 11.1107 72.6178 11.372C72.6178 11.5773 72.6925 11.75 72.8418 11.89C72.9912 12.03 73.1732 12.142 73.3878 12.226C73.6025 12.3007 73.9198 12.3987 74.3398 12.52C74.8998 12.6693 75.3572 12.8233 75.7118 12.982C76.0758 13.1314 76.3885 13.3554 76.6498 13.654C76.9112 13.9527 77.0465 14.3493 77.0558 14.844C77.0558 15.2827 76.9345 15.6747 76.6918 16.02C76.4492 16.3654 76.1038 16.636 75.6558 16.832C75.2172 17.028 74.6992 17.126 74.1018 17.126Z" fill="#1D2939"/>
+<path d="M83.4531 7.27002V17H81.8571V7.27002H83.4531Z" fill="#1D2939"/>
+<path d="M89.1605 9.16002C89.7671 9.16002 90.3085 9.28602 90.7845 9.53802C91.2698 9.79002 91.6478 10.1633 91.9185 10.658C92.1891 11.1527 92.3245 11.75 92.3245 12.45V17H90.7425V12.688C90.7425 11.9973 90.5698 11.47 90.2245 11.106C89.8791 10.7327 89.4078 10.546 88.8105 10.546C88.2131 10.546 87.7371 10.7327 87.3825 11.106C87.0371 11.47 86.8645 11.9973 86.8645 12.688V17H85.2685V9.28602H86.8645V10.168C87.1258 9.85068 87.4571 9.60335 87.8585 9.42602C88.2691 9.24868 88.7031 9.16002 89.1605 9.16002Z" fill="#1D2939"/>
+<path d="M97.3143 10.588H95.8863V17H94.2763V10.588H93.3663V9.28602H94.2763V8.74002C94.2763 7.85335 94.5096 7.20935 94.9763 6.80802C95.4523 6.39735 96.1943 6.19202 97.2023 6.19202V7.52202C96.7169 7.52202 96.3763 7.61535 96.1803 7.80202C95.9843 7.97935 95.8863 8.29202 95.8863 8.74002V9.28602H97.3143V10.588Z" fill="#1D2939"/>
+<path d="M105.519 12.954C105.519 13.2433 105.5 13.5047 105.463 13.738H99.5687C99.6154 14.354 99.844 14.8487 100.255 15.222C100.665 15.5954 101.169 15.782 101.767 15.782C102.625 15.782 103.232 15.4227 103.587 14.704H105.309C105.075 15.4133 104.651 15.9967 104.035 16.454C103.428 16.902 102.672 17.126 101.767 17.126C101.029 17.126 100.367 16.9627 99.7787 16.636C99.2 16.3 98.7427 15.8334 98.4067 15.236C98.08 14.6294 97.9167 13.9293 97.9167 13.136C97.9167 12.3427 98.0754 11.6473 98.3927 11.05C98.7194 10.4434 99.172 9.97668 99.7507 9.65002C100.339 9.32335 101.011 9.16002 101.767 9.16002C102.495 9.16002 103.143 9.31868 103.713 9.63602C104.282 9.95335 104.725 10.4014 105.043 10.98C105.36 11.5493 105.519 12.2073 105.519 12.954ZM103.853 12.45C103.843 11.862 103.633 11.3907 103.223 11.036C102.812 10.6813 102.303 10.504 101.697 10.504C101.146 10.504 100.675 10.6813 100.283 11.036C99.8907 11.3813 99.6574 11.8527 99.5827 12.45H103.853Z" fill="#1D2939"/>
+<path d="M108.405 10.406C108.639 10.014 108.947 9.71068 109.329 9.49602C109.721 9.27202 110.183 9.16002 110.715 9.16002V10.812H110.309C109.684 10.812 109.208 10.9707 108.881 11.288C108.564 11.6054 108.405 12.156 108.405 12.94V17H106.809V9.28602H108.405V10.406Z" fill="#1D2939"/>
+<path d="M118.972 12.954C118.972 13.2433 118.954 13.5047 118.916 13.738H113.022C113.069 14.354 113.298 14.8487 113.708 15.222C114.119 15.5954 114.623 15.782 115.22 15.782C116.079 15.782 116.686 15.4227 117.04 14.704H118.762C118.529 15.4133 118.104 15.9967 117.488 16.454C116.882 16.902 116.126 17.126 115.22 17.126C114.483 17.126 113.82 16.9627 113.232 16.636C112.654 16.3 112.196 15.8334 111.86 15.236C111.534 14.6294 111.37 13.9293 111.37 13.136C111.37 12.3427 111.529 11.6473 111.846 11.05C112.173 10.4434 112.626 9.97668 113.204 9.65002C113.792 9.32335 114.464 9.16002 115.22 9.16002C115.948 9.16002 116.597 9.31868 117.166 9.63602C117.736 9.95335 118.179 10.4014 118.496 10.98C118.814 11.5493 118.972 12.2073 118.972 12.954ZM117.306 12.45C117.297 11.862 117.087 11.3907 116.676 11.036C116.266 10.6813 115.757 10.504 115.15 10.504C114.6 10.504 114.128 10.6813 113.736 11.036C113.344 11.3813 113.111 11.8527 113.036 12.45H117.306Z" fill="#1D2939"/>
+<path d="M124.155 9.16002C124.762 9.16002 125.303 9.28602 125.779 9.53802C126.264 9.79002 126.642 10.1633 126.913 10.658C127.184 11.1527 127.319 11.75 127.319 12.45V17H125.737V12.688C125.737 11.9973 125.564 11.47 125.219 11.106C124.874 10.7327 124.402 10.546 123.805 10.546C123.208 10.546 122.732 10.7327 122.377 11.106C122.032 11.47 121.859 11.9973 121.859 12.688V17H120.263V9.28602H121.859V10.168C122.12 9.85068 122.452 9.60335 122.853 9.42602C123.264 9.24868 123.698 9.16002 124.155 9.16002Z" fill="#1D2939"/>
+<path d="M128.543 13.136C128.543 12.3427 128.701 11.6473 129.019 11.05C129.345 10.4434 129.793 9.97668 130.363 9.65002C130.932 9.32335 131.585 9.16002 132.323 9.16002C133.256 9.16002 134.026 9.38402 134.633 9.83202C135.249 10.2707 135.664 10.9007 135.879 11.722H134.157C134.017 11.3394 133.793 11.0407 133.485 10.826C133.177 10.6113 132.789 10.504 132.323 10.504C131.669 10.504 131.147 10.7373 130.755 11.204C130.372 11.6613 130.181 12.3053 130.181 13.136C130.181 13.9667 130.372 14.6153 130.755 15.082C131.147 15.5487 131.669 15.782 132.323 15.782C133.247 15.782 133.858 15.376 134.157 14.564H135.879C135.655 15.348 135.235 15.9733 134.619 16.44C134.003 16.8973 133.237 17.126 132.323 17.126C131.585 17.126 130.932 16.9627 130.363 16.636C129.793 16.3 129.345 15.8334 129.019 15.236C128.701 14.6294 128.543 13.9293 128.543 13.136Z" fill="#1D2939"/>
+<path d="M144.259 12.954C144.259 13.2433 144.241 13.5047 144.203 13.738H138.309C138.356 14.354 138.585 14.8487 138.995 15.222C139.406 15.5954 139.91 15.782 140.507 15.782C141.366 15.782 141.973 15.4227 142.327 14.704H144.049C143.816 15.4133 143.391 15.9967 142.775 16.454C142.169 16.902 141.413 17.126 140.507 17.126C139.77 17.126 139.107 16.9627 138.519 16.636C137.941 16.3 137.483 15.8334 137.147 15.236C136.821 14.6294 136.657 13.9293 136.657 13.136C136.657 12.3427 136.816 11.6473 137.133 11.05C137.46 10.4434 137.913 9.97668 138.491 9.65002C139.079 9.32335 139.751 9.16002 140.507 9.16002C141.235 9.16002 141.884 9.31868 142.453 9.63602C143.023 9.95335 143.466 10.4014 143.783 10.98C144.101 11.5493 144.259 12.2073 144.259 12.954ZM142.593 12.45C142.584 11.862 142.374 11.3907 141.963 11.036C141.553 10.6813 141.044 10.504 140.437 10.504C139.887 10.504 139.415 10.6813 139.023 11.036C138.631 11.3813 138.398 11.8527 138.323 12.45H142.593Z" fill="#1D2939"/>
+</g>
+</g>
+<defs>
+<linearGradient id="paint0_linear_9866_6170" x1="2.15214" y1="24.3018" x2="21.2921" y2="0.0988218" gradientUnits="userSpaceOnUse">
+<stop stop-color="#E9A85E"/>
+<stop offset="1" stop-color="#F52B76"/>
+</linearGradient>
+<linearGradient id="paint1_linear_9866_6170" x1="2.06269" y1="24.2294" x2="21.2027" y2="0.028252" gradientUnits="userSpaceOnUse">
+<stop stop-color="#E9A85E"/>
+<stop offset="1" stop-color="#F52B76"/>
+</linearGradient>
+<linearGradient id="paint2_linear_9866_6170" x1="-0.613606" y1="3.843" x2="21.4449" y2="18.7258" gradientUnits="userSpaceOnUse">
+<stop stop-color="#6A0CF5"/>
+<stop offset="1" stop-color="#AB66F3"/>
+</linearGradient>
+<clipPath id="clip0_9866_6170">
+<rect width="152" height="24" fill="white"/>
+</clipPath>
+</defs>
+</svg>

web/app/components/base/icons/assets/public/llm/xorbits-inference.svg

@@ -0,0 +1,24 @@
+<svg width="24" height="24" viewBox="0 0 24 24" fill="none" xmlns="http://www.w3.org/2000/svg">
+<g id="Xorbits Square" clip-path="url(#clip0_9850_26870)">
+<path id="Vector" d="M8.00391 12.3124C8.69334 13.0754 9.47526 13.7494 10.3316 14.3188C11.0667 14.8105 11.8509 15.2245 12.6716 15.5541C14.1617 14.1465 15.3959 12.4907 16.3192 10.6606L21.7051 0L12.3133 7.38353C10.5832 8.74456 9.12178 10.416 8.00391 12.3124Z" fill="url(#paint0_linear_9850_26870)"/>
+<path id="Vector_2" d="M7.23504 18.9512C6.56092 18.5012 5.92386 18.0265 5.3221 17.5394L2.06445 24L7.91975 19.3959C7.69034 19.2494 7.46092 19.103 7.23504 18.9512Z" fill="url(#paint1_linear_9850_26870)"/>
+<path id="Vector_3" d="M19.3161 8.57474C21.0808 10.9147 21.5961 13.5159 20.3996 15.3053C18.6526 17.9189 13.9161 17.8183 9.82024 15.0812C5.72435 12.3441 3.82024 8.0065 5.56729 5.39297C6.76377 3.60356 9.36318 3.0865 12.2008 3.81886C7.29318 1.73474 2.62376 1.94121 0.813177 4.64474C-1.45976 8.04709 1.64435 14.1177 7.74494 18.1889C13.8455 22.26 20.6361 22.8124 22.9091 19.4118C24.7179 16.703 23.1173 12.3106 19.3161 8.57474Z" fill="url(#paint2_linear_9850_26870)"/>
+</g>
+<defs>
+<linearGradient id="paint0_linear_9850_26870" x1="2.15214" y1="24.3018" x2="21.2921" y2="0.0988218" gradientUnits="userSpaceOnUse">
+<stop stop-color="#E9A85E"/>
+<stop offset="1" stop-color="#F52B76"/>
+</linearGradient>
+<linearGradient id="paint1_linear_9850_26870" x1="2.06269" y1="24.2294" x2="21.2027" y2="0.028252" gradientUnits="userSpaceOnUse">
+<stop stop-color="#E9A85E"/>
+<stop offset="1" stop-color="#F52B76"/>
+</linearGradient>
+<linearGradient id="paint2_linear_9850_26870" x1="-0.613606" y1="3.843" x2="21.4449" y2="18.7258" gradientUnits="userSpaceOnUse">
+<stop stop-color="#6A0CF5"/>
+<stop offset="1" stop-color="#AB66F3"/>
+</linearGradient>
+<clipPath id="clip0_9850_26870">
+<rect width="24" height="24" fill="white"/>
+</clipPath>
+</defs>
+</svg>

web/app/components/base/icons/src/public/llm/Openllm.tsx

@@ -0,0 +1,14 @@
+// GENERATE BY script
+// DON NOT EDIT IT MANUALLY
+
+import * as React from 'react'
+import data from './Openllm.json'
+import IconBase from '@/app/components/base/icons/IconBase'
+import type { IconBaseProps, IconData } from '@/app/components/base/icons/IconBase'
+
+const Icon = React.forwardRef<React.MutableRefObject<SVGElement>, Omit<IconBaseProps, 'data'>>((
+  props,
+  ref,
+) => <IconBase {...props} ref={ref} data={data as IconData} />)
+
+export default Icon

web/app/components/base/icons/src/public/llm/OpenllmText.tsx

@@ -0,0 +1,14 @@
+// GENERATE BY script
+// DON NOT EDIT IT MANUALLY
+
+import * as React from 'react'
+import data from './OpenllmText.json'
+import IconBase from '@/app/components/base/icons/IconBase'
+import type { IconBaseProps, IconData } from '@/app/components/base/icons/IconBase'
+
+const Icon = React.forwardRef<React.MutableRefObject<SVGElement>, Omit<IconBaseProps, 'data'>>((
+  props,
+  ref,
+) => <IconBase {...props} ref={ref} data={data as IconData} />)
+
+export default Icon

web/app/components/base/icons/src/public/llm/XorbitsInference.json

@@ -0,0 +1,176 @@
+{
+	"icon": {
+		"type": "element",
+		"isRootNode": true,
+		"name": "svg",
+		"attributes": {
+			"width": "24",
+			"height": "24",
+			"viewBox": "0 0 24 24",
+			"fill": "none",
+			"xmlns": "http://www.w3.org/2000/svg"
+		},
+		"children": [
+			{
+				"type": "element",
+				"name": "g",
+				"attributes": {
+					"id": "Xorbits Square",
+					"clip-path": "url(#clip0_9850_26870)"
+				},
+				"children": [
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"id": "Vector",
+							"d": "M8.00391 12.3124C8.69334 13.0754 9.47526 13.7494 10.3316 14.3188C11.0667 14.8105 11.8509 15.2245 12.6716 15.5541C14.1617 14.1465 15.3959 12.4907 16.3192 10.6606L21.7051 0L12.3133 7.38353C10.5832 8.74456 9.12178 10.416 8.00391 12.3124Z",
+							"fill": "url(#paint0_linear_9850_26870)"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"id": "Vector_2",
+							"d": "M7.23504 18.9512C6.56092 18.5012 5.92386 18.0265 5.3221 17.5394L2.06445 24L7.91975 19.3959C7.69034 19.2494 7.46092 19.103 7.23504 18.9512Z",
+							"fill": "url(#paint1_linear_9850_26870)"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"id": "Vector_3",
+							"d": "M19.3161 8.57474C21.0808 10.9147 21.5961 13.5159 20.3996 15.3053C18.6526 17.9189 13.9161 17.8183 9.82024 15.0812C5.72435 12.3441 3.82024 8.0065 5.56729 5.39297C6.76377 3.60356 9.36318 3.0865 12.2008 3.81886C7.29318 1.73474 2.62376 1.94121 0.813177 4.64474C-1.45976 8.04709 1.64435 14.1177 7.74494 18.1889C13.8455 22.26 20.6361 22.8124 22.9091 19.4118C24.7179 16.703 23.1173 12.3106 19.3161 8.57474Z",
+							"fill": "url(#paint2_linear_9850_26870)"
+						},
+						"children": []
+					}
+				]
+			},
+			{
+				"type": "element",
+				"name": "defs",
+				"attributes": {},
+				"children": [
+					{
+						"type": "element",
+						"name": "linearGradient",
+						"attributes": {
+							"id": "paint0_linear_9850_26870",
+							"x1": "2.15214",
+							"y1": "24.3018",
+							"x2": "21.2921",
+							"y2": "0.0988218",
+							"gradientUnits": "userSpaceOnUse"
+						},
+						"children": [
+							{
+								"type": "element",
+								"name": "stop",
+								"attributes": {
+									"stop-color": "#E9A85E"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "stop",
+								"attributes": {
+									"offset": "1",
+									"stop-color": "#F52B76"
+								},
+								"children": []
+							}
+						]
+					},
+					{
+						"type": "element",
+						"name": "linearGradient",
+						"attributes": {
+							"id": "paint1_linear_9850_26870",
+							"x1": "2.06269",
+							"y1": "24.2294",
+							"x2": "21.2027",
+							"y2": "0.028252",
+							"gradientUnits": "userSpaceOnUse"
+						},
+						"children": [
+							{
+								"type": "element",
+								"name": "stop",
+								"attributes": {
+									"stop-color": "#E9A85E"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "stop",
+								"attributes": {
+									"offset": "1",
+									"stop-color": "#F52B76"
+								},
+								"children": []
+							}
+						]
+					},
+					{
+						"type": "element",
+						"name": "linearGradient",
+						"attributes": {
+							"id": "paint2_linear_9850_26870",
+							"x1": "-0.613606",
+							"y1": "3.843",
+							"x2": "21.4449",
+							"y2": "18.7258",
+							"gradientUnits": "userSpaceOnUse"
+						},
+						"children": [
+							{
+								"type": "element",
+								"name": "stop",
+								"attributes": {
+									"stop-color": "#6A0CF5"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "stop",
+								"attributes": {
+									"offset": "1",
+									"stop-color": "#AB66F3"
+								},
+								"children": []
+							}
+						]
+					},
+					{
+						"type": "element",
+						"name": "clipPath",
+						"attributes": {
+							"id": "clip0_9850_26870"
+						},
+						"children": [
+							{
+								"type": "element",
+								"name": "rect",
+								"attributes": {
+									"width": "24",
+									"height": "24",
+									"fill": "white"
+								},
+								"children": []
+							}
+						]
+					}
+				]
+			}
+		]
+	},
+	"name": "XorbitsInference"
+}

web/app/components/base/icons/src/public/llm/XorbitsInference.tsx

@@ -0,0 +1,14 @@
+// GENERATE BY script
+// DON NOT EDIT IT MANUALLY
+
+import * as React from 'react'
+import data from './XorbitsInference.json'
+import IconBase from '@/app/components/base/icons/IconBase'
+import type { IconBaseProps, IconData } from '@/app/components/base/icons/IconBase'
+
+const Icon = React.forwardRef<React.MutableRefObject<SVGElement>, Omit<IconBaseProps, 'data'>>((
+  props,
+  ref,
+) => <IconBase {...props} ref={ref} data={data as IconData} />)
+
+export default Icon

web/app/components/base/icons/src/public/llm/XorbitsInferenceText.json

@@ -0,0 +1,329 @@
+{
+	"icon": {
+		"type": "element",
+		"isRootNode": true,
+		"name": "svg",
+		"attributes": {
+			"width": "152",
+			"height": "24",
+			"viewBox": "0 0 152 24",
+			"fill": "none",
+			"xmlns": "http://www.w3.org/2000/svg"
+		},
+		"children": [
+			{
+				"type": "element",
+				"name": "g",
+				"attributes": {
+					"id": "xorbits 1",
+					"clip-path": "url(#clip0_9866_6170)"
+				},
+				"children": [
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"id": "Vector",
+							"d": "M8.00391 12.3124C8.69334 13.0754 9.47526 13.7494 10.3316 14.3188C11.0667 14.8105 11.8509 15.2245 12.6716 15.5541C14.1617 14.1465 15.3959 12.4907 16.3192 10.6606L21.7051 0L12.3133 7.38353C10.5832 8.74456 9.12178 10.416 8.00391 12.3124Z",
+							"fill": "url(#paint0_linear_9866_6170)"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"id": "Vector_2",
+							"d": "M7.23504 18.9512C6.56092 18.5012 5.92386 18.0265 5.3221 17.5394L2.06445 24L7.91975 19.3959C7.69034 19.2494 7.46092 19.103 7.23504 18.9512Z",
+							"fill": "url(#paint1_linear_9866_6170)"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"id": "Vector_3",
+							"d": "M19.3161 8.57474C21.0808 10.9147 21.5961 13.5159 20.3996 15.3053C18.6526 17.9189 13.9161 17.8183 9.82024 15.0812C5.72435 12.3441 3.82024 8.0065 5.56729 5.39297C6.76377 3.60356 9.36318 3.0865 12.2008 3.81886C7.29318 1.73474 2.62376 1.94121 0.813177 4.64474C-1.45976 8.04709 1.64435 14.1177 7.74494 18.1889C13.8455 22.26 20.6361 22.8124 22.9091 19.4118C24.7179 16.703 23.1173 12.3106 19.3161 8.57474Z",
+							"fill": "url(#paint2_linear_9866_6170)"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "g",
+						"attributes": {
+							"id": "Xorbits Inference"
+						},
+						"children": [
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M35.5162 12.142L38.5402 17H36.7482L34.5502 13.472L32.4922 17H30.7142L33.7382 12.142L30.7002 7.27002H32.4922L34.7042 10.826L36.7762 7.27002H38.5542L35.5162 12.142Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M43.3584 17.126C42.6304 17.126 41.9724 16.9627 41.3844 16.636C40.7964 16.3 40.3344 15.8334 39.9984 15.236C39.6624 14.6294 39.4944 13.9293 39.4944 13.136C39.4944 12.352 39.6671 11.6567 40.0124 11.05C40.3577 10.4434 40.8291 9.97668 41.4264 9.65002C42.0237 9.32335 42.6911 9.16002 43.4284 9.16002C44.1657 9.16002 44.8331 9.32335 45.4304 9.65002C46.0277 9.97668 46.4991 10.4434 46.8444 11.05C47.1897 11.6567 47.3624 12.352 47.3624 13.136C47.3624 13.92 47.185 14.6154 46.8304 15.222C46.4757 15.8287 45.9904 16.3 45.3744 16.636C44.7677 16.9627 44.0957 17.126 43.3584 17.126ZM43.3584 15.74C43.769 15.74 44.1517 15.642 44.5064 15.446C44.8704 15.25 45.1644 14.956 45.3884 14.564C45.6124 14.172 45.7244 13.696 45.7244 13.136C45.7244 12.576 45.6171 12.1047 45.4024 11.722C45.1877 11.33 44.9031 11.036 44.5484 10.84C44.1937 10.644 43.8111 10.546 43.4004 10.546C42.9897 10.546 42.607 10.644 42.2524 10.84C41.9071 11.036 41.6317 11.33 41.4264 11.722C41.221 12.1047 41.1184 12.576 41.1184 13.136C41.1184 13.9667 41.3284 14.6107 41.7484 15.068C42.1777 15.516 42.7144 15.74 43.3584 15.74Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M50.2561 10.406C50.4895 10.014 50.7974 9.71068 51.1801 9.49602C51.5721 9.27202 52.0341 9.16002 52.5661 9.16002V10.812H52.1601C51.5348 10.812 51.0588 10.9707 50.7321 11.288C50.4148 11.6054 50.2561 12.156 50.2561 12.94V17H48.6601V9.28602H50.2561V10.406Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M55.3492 10.434C55.6198 10.0607 55.9885 9.75735 56.4552 9.52402C56.9312 9.28135 57.4585 9.16002 58.0372 9.16002C58.7185 9.16002 59.3345 9.32335 59.8852 9.65002C60.4358 9.97668 60.8698 10.4434 61.1872 11.05C61.5045 11.6473 61.6632 12.3333 61.6632 13.108C61.6632 13.8827 61.5045 14.578 61.1872 15.194C60.8698 15.8007 60.4312 16.2767 59.8712 16.622C59.3205 16.958 58.7092 17.126 58.0372 17.126C57.4398 17.126 56.9078 17.0093 56.4412 16.776C55.9838 16.5427 55.6198 16.244 55.3492 15.88V17H53.7532V6.64002H55.3492V10.434ZM60.0392 13.108C60.0392 12.576 59.9272 12.1187 59.7032 11.736C59.4885 11.344 59.1992 11.05 58.8352 10.854C58.4805 10.6487 58.0978 10.546 57.6872 10.546C57.2858 10.546 56.9032 10.6487 56.5392 10.854C56.1845 11.0594 55.8952 11.358 55.6712 11.75C55.4565 12.142 55.3492 12.604 55.3492 13.136C55.3492 13.668 55.4565 14.1347 55.6712 14.536C55.8952 14.928 56.1845 15.2267 56.5392 15.432C56.9032 15.6374 57.2858 15.74 57.6872 15.74C58.0978 15.74 58.4805 15.6374 58.8352 15.432C59.1992 15.2174 59.4885 14.9093 59.7032 14.508C59.9272 14.1067 60.0392 13.64 60.0392 13.108Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M63.7734 8.26402C63.4841 8.26402 63.2414 8.16602 63.0454 7.97002C62.8494 7.77402 62.7514 7.53135 62.7514 7.24202C62.7514 6.95268 62.8494 6.71002 63.0454 6.51402C63.2414 6.31802 63.4841 6.22002 63.7734 6.22002C64.0534 6.22002 64.2914 6.31802 64.4874 6.51402C64.6834 6.71002 64.7814 6.95268 64.7814 7.24202C64.7814 7.53135 64.6834 7.77402 64.4874 7.97002C64.2914 8.16602 64.0534 8.26402 63.7734 8.26402ZM64.5574 9.28602V17H62.9614V9.28602H64.5574Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M68.2348 10.588V14.858C68.2348 15.1474 68.3002 15.3573 68.4309 15.488C68.5709 15.6093 68.8042 15.67 69.1308 15.67H70.1109V17H68.8508C68.1322 17 67.5815 16.832 67.1988 16.496C66.8162 16.16 66.6248 15.614 66.6248 14.858V10.588H65.7148V9.28602H66.6248V7.36802H68.2348V9.28602H70.1109V10.588H68.2348Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M74.1018 17.126C73.4952 17.126 72.9492 17.0187 72.4638 16.804C71.9878 16.58 71.6098 16.2813 71.3298 15.908C71.0498 15.5253 70.9005 15.1007 70.8818 14.634H72.5338C72.5618 14.9607 72.7158 15.236 72.9958 15.46C73.2852 15.6747 73.6445 15.782 74.0738 15.782C74.5218 15.782 74.8672 15.698 75.1098 15.53C75.3618 15.3527 75.4878 15.1287 75.4878 14.858C75.4878 14.5687 75.3478 14.354 75.0678 14.214C74.7972 14.074 74.3632 13.92 73.7658 13.752C73.1872 13.5933 72.7158 13.4394 72.3518 13.29C71.9878 13.1407 71.6705 12.912 71.3998 12.604C71.1385 12.296 71.0078 11.89 71.0078 11.386C71.0078 10.9753 71.1292 10.602 71.3718 10.266C71.6145 9.92068 71.9598 9.65002 72.4078 9.45402C72.8652 9.25802 73.3878 9.16002 73.9758 9.16002C74.8532 9.16002 75.5578 9.38402 76.0898 9.83202C76.6312 10.2707 76.9205 10.8727 76.9578 11.638H75.3618C75.3338 11.2927 75.1938 11.0173 74.9418 10.812C74.6898 10.6067 74.3492 10.504 73.9198 10.504C73.4998 10.504 73.1778 10.5833 72.9538 10.742C72.7298 10.9007 72.6178 11.1107 72.6178 11.372C72.6178 11.5773 72.6925 11.75 72.8418 11.89C72.9912 12.03 73.1732 12.142 73.3878 12.226C73.6025 12.3007 73.9198 12.3987 74.3398 12.52C74.8998 12.6693 75.3572 12.8233 75.7118 12.982C76.0758 13.1314 76.3885 13.3554 76.6498 13.654C76.9112 13.9527 77.0465 14.3493 77.0558 14.844C77.0558 15.2827 76.9345 15.6747 76.6918 16.02C76.4492 16.3654 76.1038 16.636 75.6558 16.832C75.2172 17.028 74.6992 17.126 74.1018 17.126Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M83.4531 7.27002V17H81.8571V7.27002H83.4531Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M89.1605 9.16002C89.7671 9.16002 90.3085 9.28602 90.7845 9.53802C91.2698 9.79002 91.6478 10.1633 91.9185 10.658C92.1891 11.1527 92.3245 11.75 92.3245 12.45V17H90.7425V12.688C90.7425 11.9973 90.5698 11.47 90.2245 11.106C89.8791 10.7327 89.4078 10.546 88.8105 10.546C88.2131 10.546 87.7371 10.7327 87.3825 11.106C87.0371 11.47 86.8645 11.9973 86.8645 12.688V17H85.2685V9.28602H86.8645V10.168C87.1258 9.85068 87.4571 9.60335 87.8585 9.42602C88.2691 9.24868 88.7031 9.16002 89.1605 9.16002Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M97.3143 10.588H95.8863V17H94.2763V10.588H93.3663V9.28602H94.2763V8.74002C94.2763 7.85335 94.5096 7.20935 94.9763 6.80802C95.4523 6.39735 96.1943 6.19202 97.2023 6.19202V7.52202C96.7169 7.52202 96.3763 7.61535 96.1803 7.80202C95.9843 7.97935 95.8863 8.29202 95.8863 8.74002V9.28602H97.3143V10.588Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M105.519 12.954C105.519 13.2433 105.5 13.5047 105.463 13.738H99.5687C99.6154 14.354 99.844 14.8487 100.255 15.222C100.665 15.5954 101.169 15.782 101.767 15.782C102.625 15.782 103.232 15.4227 103.587 14.704H105.309C105.075 15.4133 104.651 15.9967 104.035 16.454C103.428 16.902 102.672 17.126 101.767 17.126C101.029 17.126 100.367 16.9627 99.7787 16.636C99.2 16.3 98.7427 15.8334 98.4067 15.236C98.08 14.6294 97.9167 13.9293 97.9167 13.136C97.9167 12.3427 98.0754 11.6473 98.3927 11.05C98.7194 10.4434 99.172 9.97668 99.7507 9.65002C100.339 9.32335 101.011 9.16002 101.767 9.16002C102.495 9.16002 103.143 9.31868 103.713 9.63602C104.282 9.95335 104.725 10.4014 105.043 10.98C105.36 11.5493 105.519 12.2073 105.519 12.954ZM103.853 12.45C103.843 11.862 103.633 11.3907 103.223 11.036C102.812 10.6813 102.303 10.504 101.697 10.504C101.146 10.504 100.675 10.6813 100.283 11.036C99.8907 11.3813 99.6574 11.8527 99.5827 12.45H103.853Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M108.405 10.406C108.639 10.014 108.947 9.71068 109.329 9.49602C109.721 9.27202 110.183 9.16002 110.715 9.16002V10.812H110.309C109.684 10.812 109.208 10.9707 108.881 11.288C108.564 11.6054 108.405 12.156 108.405 12.94V17H106.809V9.28602H108.405V10.406Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M118.972 12.954C118.972 13.2433 118.954 13.5047 118.916 13.738H113.022C113.069 14.354 113.298 14.8487 113.708 15.222C114.119 15.5954 114.623 15.782 115.22 15.782C116.079 15.782 116.686 15.4227 117.04 14.704H118.762C118.529 15.4133 118.104 15.9967 117.488 16.454C116.882 16.902 116.126 17.126 115.22 17.126C114.483 17.126 113.82 16.9627 113.232 16.636C112.654 16.3 112.196 15.8334 111.86 15.236C111.534 14.6294 111.37 13.9293 111.37 13.136C111.37 12.3427 111.529 11.6473 111.846 11.05C112.173 10.4434 112.626 9.97668 113.204 9.65002C113.792 9.32335 114.464 9.16002 115.22 9.16002C115.948 9.16002 116.597 9.31868 117.166 9.63602C117.736 9.95335 118.179 10.4014 118.496 10.98C118.814 11.5493 118.972 12.2073 118.972 12.954ZM117.306 12.45C117.297 11.862 117.087 11.3907 116.676 11.036C116.266 10.6813 115.757 10.504 115.15 10.504C114.6 10.504 114.128 10.6813 113.736 11.036C113.344 11.3813 113.111 11.8527 113.036 12.45H117.306Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M124.155 9.16002C124.762 9.16002 125.303 9.28602 125.779 9.53802C126.264 9.79002 126.642 10.1633 126.913 10.658C127.184 11.1527 127.319 11.75 127.319 12.45V17H125.737V12.688C125.737 11.9973 125.564 11.47 125.219 11.106C124.874 10.7327 124.402 10.546 123.805 10.546C123.208 10.546 122.732 10.7327 122.377 11.106C122.032 11.47 121.859 11.9973 121.859 12.688V17H120.263V9.28602H121.859V10.168C122.12 9.85068 122.452 9.60335 122.853 9.42602C123.264 9.24868 123.698 9.16002 124.155 9.16002Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M128.543 13.136C128.543 12.3427 128.701 11.6473 129.019 11.05C129.345 10.4434 129.793 9.97668 130.363 9.65002C130.932 9.32335 131.585 9.16002 132.323 9.16002C133.256 9.16002 134.026 9.38402 134.633 9.83202C135.249 10.2707 135.664 10.9007 135.879 11.722H134.157C134.017 11.3394 133.793 11.0407 133.485 10.826C133.177 10.6113 132.789 10.504 132.323 10.504C131.669 10.504 131.147 10.7373 130.755 11.204C130.372 11.6613 130.181 12.3053 130.181 13.136C130.181 13.9667 130.372 14.6153 130.755 15.082C131.147 15.5487 131.669 15.782 132.323 15.782C133.247 15.782 133.858 15.376 134.157 14.564H135.879C135.655 15.348 135.235 15.9733 134.619 16.44C134.003 16.8973 133.237 17.126 132.323 17.126C131.585 17.126 130.932 16.9627 130.363 16.636C129.793 16.3 129.345 15.8334 129.019 15.236C128.701 14.6294 128.543 13.9293 128.543 13.136Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "path",
+								"attributes": {
+									"d": "M144.259 12.954C144.259 13.2433 144.241 13.5047 144.203 13.738H138.309C138.356 14.354 138.585 14.8487 138.995 15.222C139.406 15.5954 139.91 15.782 140.507 15.782C141.366 15.782 141.973 15.4227 142.327 14.704H144.049C143.816 15.4133 143.391 15.9967 142.775 16.454C142.169 16.902 141.413 17.126 140.507 17.126C139.77 17.126 139.107 16.9627 138.519 16.636C137.941 16.3 137.483 15.8334 137.147 15.236C136.821 14.6294 136.657 13.9293 136.657 13.136C136.657 12.3427 136.816 11.6473 137.133 11.05C137.46 10.4434 137.913 9.97668 138.491 9.65002C139.079 9.32335 139.751 9.16002 140.507 9.16002C141.235 9.16002 141.884 9.31868 142.453 9.63602C143.023 9.95335 143.466 10.4014 143.783 10.98C144.101 11.5493 144.259 12.2073 144.259 12.954ZM142.593 12.45C142.584 11.862 142.374 11.3907 141.963 11.036C141.553 10.6813 141.044 10.504 140.437 10.504C139.887 10.504 139.415 10.6813 139.023 11.036C138.631 11.3813 138.398 11.8527 138.323 12.45H142.593Z",
+									"fill": "#1D2939"
+								},
+								"children": []
+							}
+						]
+					}
+				]
+			},
+			{
+				"type": "element",
+				"name": "defs",
+				"attributes": {},
+				"children": [
+					{
+						"type": "element",
+						"name": "linearGradient",
+						"attributes": {
+							"id": "paint0_linear_9866_6170",
+							"x1": "2.15214",
+							"y1": "24.3018",
+							"x2": "21.2921",
+							"y2": "0.0988218",
+							"gradientUnits": "userSpaceOnUse"
+						},
+						"children": [
+							{
+								"type": "element",
+								"name": "stop",
+								"attributes": {
+									"stop-color": "#E9A85E"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "stop",
+								"attributes": {
+									"offset": "1",
+									"stop-color": "#F52B76"
+								},
+								"children": []
+							}
+						]
+					},
+					{
+						"type": "element",
+						"name": "linearGradient",
+						"attributes": {
+							"id": "paint1_linear_9866_6170",
+							"x1": "2.06269",
+							"y1": "24.2294",
+							"x2": "21.2027",
+							"y2": "0.028252",
+							"gradientUnits": "userSpaceOnUse"
+						},
+						"children": [
+							{
+								"type": "element",
+								"name": "stop",
+								"attributes": {
+									"stop-color": "#E9A85E"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "stop",
+								"attributes": {
+									"offset": "1",
+									"stop-color": "#F52B76"
+								},
+								"children": []
+							}
+						]
+					},
+					{
+						"type": "element",
+						"name": "linearGradient",
+						"attributes": {
+							"id": "paint2_linear_9866_6170",
+							"x1": "-0.613606",
+							"y1": "3.843",
+							"x2": "21.4449",
+							"y2": "18.7258",
+							"gradientUnits": "userSpaceOnUse"
+						},
+						"children": [
+							{
+								"type": "element",
+								"name": "stop",
+								"attributes": {
+									"stop-color": "#6A0CF5"
+								},
+								"children": []
+							},
+							{
+								"type": "element",
+								"name": "stop",
+								"attributes": {
+									"offset": "1",
+									"stop-color": "#AB66F3"
+								},
+								"children": []
+							}
+						]
+					},
+					{
+						"type": "element",
+						"name": "clipPath",
+						"attributes": {
+							"id": "clip0_9866_6170"
+						},
+						"children": [
+							{
+								"type": "element",
+								"name": "rect",
+								"attributes": {
+									"width": "152",
+									"height": "24",
+									"fill": "white"
+								},
+								"children": []
+							}
+						]
+					}
+				]
+			}
+		]
+	},
+	"name": "XorbitsInferenceText"
+}

web/app/components/base/icons/src/public/llm/XorbitsInferenceText.tsx

@@ -0,0 +1,14 @@
+// GENERATE BY script
+// DON NOT EDIT IT MANUALLY
+
+import * as React from 'react'
+import data from './XorbitsInferenceText.json'
+import IconBase from '@/app/components/base/icons/IconBase'
+import type { IconBaseProps, IconData } from '@/app/components/base/icons/IconBase'
+
+const Icon = React.forwardRef<React.MutableRefObject<SVGElement>, Omit<IconBaseProps, 'data'>>((
+  props,
+  ref,
+) => <IconBase {...props} ref={ref} data={data as IconData} />)
+
+export default Icon

web/app/components/base/icons/src/public/llm/index.ts

@@ -21,5 +21,9 @@ export { default as OpenaiGreen } from './OpenaiGreen'
 export { default as OpenaiText } from './OpenaiText'
 export { default as OpenaiTransparent } from './OpenaiTransparent'
 export { default as OpenaiViolet } from './OpenaiViolet'
+export { default as OpenllmText } from './OpenllmText'
+export { default as Openllm } from './Openllm'
 export { default as ReplicateText } from './ReplicateText'
 export { default as Replicate } from './Replicate'
+export { default as XorbitsInferenceText } from './XorbitsInferenceText'
+export { default as XorbitsInference } from './XorbitsInference'

web/app/components/base/icons/src/vender/line/general/DotsHorizontal.json

@@ -68,4 +68,4 @@
 		]
 	},
 	"name": "DotsHorizontal"
-}
+}

web/app/components/base/loading/style.css

@@ -1,5 +1,5 @@
 .spin-animation path {
-    animation: custom 1s linear infinite;
+    animation: custom 2s linear infinite;
 }
 
 @keyframes custom {
@@ -29,13 +29,13 @@
 }
 
 .spin-animation path:nth-child(2) {
-    animation-delay: 0.25s;
-}
-
-.spin-animation path:nth-child(3) {
     animation-delay: 0.5s;
 }
 
-.spin-animation path:nth-child(4) {
+.spin-animation path:nth-child(3) {
     animation-delay: 1s;
 }
+
+.spin-animation path:nth-child(4) {
+    animation-delay: 2s;
+}

web/app/components/base/portal-to-follow-elem/index.tsx

@@ -1,9 +1,10 @@
 'use client'
 import { useBoolean } from 'ahooks'
-import React, { FC, useEffect, useState, useRef } from 'react'
+import React, { useEffect, useRef, useState } from 'react'
+import type { FC } from 'react'
 import { createRoot } from 'react-dom/client'
 
-export interface IPortalToFollowElementProps {
+type IPortalToFollowElementProps = {
   portalElem: React.ReactNode
   children: React.ReactNode
   controlShow?: number
@@ -14,44 +15,42 @@ const PortalToFollowElement: FC<IPortalToFollowElementProps> = ({
   portalElem,
   children,
   controlShow,
-  controlHide
+  controlHide,
 }) => {
   const [isShowContent, { setTrue: showContent, setFalse: hideContent, toggle: toggleContent }] = useBoolean(false)
   const [wrapElem, setWrapElem] = useState<HTMLDivElement | null>(null)
 
   useEffect(() => {
-    if (controlShow) {
+    if (controlShow)
       showContent()
-    }
   }, [controlShow])
 
   useEffect(() => {
-    if (controlHide) {
+    if (controlHide)
       hideContent()
-    }
   }, [controlHide])
 
   // todo use click outside hidden
-  const triggerElemRef = useRef<HTMLElement>(null)
+  const triggerElemRef = useRef<HTMLDivElement>(null)
 
   const calLoc = () => {
     const triggerElem = triggerElemRef.current
     if (!triggerElem) {
       return {
-        display: 'none'
+        display: 'none',
       }
     }
     const {
       left: triggerLeft,
       top: triggerTop,
-      height
-    } = triggerElem.getBoundingClientRect();
+      height,
+    } = triggerElem.getBoundingClientRect()
 
     return {
       position: 'fixed',
       left: triggerLeft,
       top: triggerTop + height,
-      zIndex: 999
+      zIndex: 999,
     }
   }
 
@@ -63,19 +62,20 @@ const PortalToFollowElement: FC<IPortalToFollowElementProps> = ({
       root.render(
         <div style={style as React.CSSProperties}>
           {portalElem}
-        </div>
+        </div>,
       )
       document.body.appendChild(holder)
       setWrapElem(holder)
       console.log(holder)
-    } else {
+    }
+    else {
       wrapElem?.remove?.()
       setWrapElem(null)
     }
   }, [isShowContent])
 
   return (
-    <div ref={triggerElemRef as any} onClick={toggleContent}>
+    <div ref={triggerElemRef as React.RefObject<HTMLDivElement>} onClick={toggleContent}>
       {children}
     </div>
   )