fix: gen-icon script phantom deps

Co-authored-by: cuihz <cuihz@knowbox.cn>

.devcontainer/post_create_command.sh

@@ -1,11 +1,12 @@
 #!/bin/bash
 
-cd web && npm install
+npm add -g pnpm@9.12.2
+cd web && pnpm install
 pipx install poetry
 
 echo 'alias start-api="cd /workspaces/dify/api && poetry run python -m flask run --host 0.0.0.0 --port=5001 --debug"' >> ~/.bashrc
 echo 'alias start-worker="cd /workspaces/dify/api && poetry run python -m celery -A app.celery worker -P gevent -c 1 --loglevel INFO -Q dataset,generation,mail,ops_trace,app_deletion"' >> ~/.bashrc
-echo 'alias start-web="cd /workspaces/dify/web && npm run dev"' >> ~/.bashrc
+echo 'alias start-web="cd /workspaces/dify/web && pnpm dev"' >> ~/.bashrc
 echo 'alias start-containers="cd /workspaces/dify/docker && docker-compose -f docker-compose.middleware.yaml -p dify up -d"' >> ~/.bashrc
 
-source /home/vscode/.bashrc
+source /home/vscode/.bashrc

.github/workflows/api-tests.yml

@@ -27,18 +27,17 @@ jobs:
       - name: Checkout code
         uses: actions/checkout@v4
 
-      - name: Install Poetry
-        uses: abatilo/actions-poetry@v3
-
       - name: Set up Python ${{ matrix.python-version }}
         uses: actions/setup-python@v5
         with:
           python-version: ${{ matrix.python-version }}
-          cache: 'poetry'
           cache-dependency-path: |
             api/pyproject.toml
             api/poetry.lock
 
+      - name: Install Poetry
+        uses: abatilo/actions-poetry@v3
+
       - name: Check Poetry lockfile
         run: |
           poetry check -C api --lock

.github/workflows/db-migration-test.yml

@@ -23,18 +23,17 @@ jobs:
       - name: Checkout code
         uses: actions/checkout@v4
 
-      - name: Install Poetry
-        uses: abatilo/actions-poetry@v3
-
       - name: Set up Python ${{ matrix.python-version }}
         uses: actions/setup-python@v5
         with:
           python-version: ${{ matrix.python-version }}
-          cache: 'poetry'
           cache-dependency-path: |
             api/pyproject.toml
             api/poetry.lock
 
+      - name: Install Poetry
+        uses: abatilo/actions-poetry@v3
+
       - name: Install dependencies
         run: poetry install -C api
 

.github/workflows/style.yml

@@ -24,15 +24,16 @@ jobs:
         with:
           files: api/**
 
-      - name: Install Poetry
-        uses: abatilo/actions-poetry@v3
-
       - name: Set up Python
         uses: actions/setup-python@v5
         if: steps.changed-files.outputs.any_changed == 'true'
         with:
           python-version: '3.10'
 
+      - name: Install Poetry
+        if: steps.changed-files.outputs.any_changed == 'true'
+        uses: abatilo/actions-poetry@v3
+
       - name: Python dependencies
         if: steps.changed-files.outputs.any_changed == 'true'
         run: poetry install -C api --only lint

.github/workflows/translate-i18n-base-on-english.yml

@@ -42,7 +42,7 @@ jobs:
 
       - name: Run npm script
         if: env.FILES_CHANGED == 'true'
-        run: npm run auto-gen-i18n
+        run: pnpm run auto-gen-i18n
 
       - name: Create Pull Request
         if: env.FILES_CHANGED == 'true'

.gitignore

@@ -175,6 +175,8 @@ docker/volumes/pgvector/data/*
 docker/volumes/pgvecto_rs/data/*
 
 docker/nginx/conf.d/default.conf
+docker/nginx/ssl/*
+!docker/nginx/ssl/.gitkeep
 docker/middleware.env
 
 sdks/python-client/build
@@ -187,4 +189,7 @@ pyrightconfig.json
 api/.vscode
 
 .idea/
-.vscode
+.vscode
+
+# pnpm
+/.pnpm-store

LICENSE

@@ -6,8 +6,9 @@ Dify is licensed under the Apache License 2.0, with the following additional con
 
 a. Multi-tenant service: Unless explicitly authorized by Dify in writing, you may not use the Dify source code to operate a multi-tenant environment. 
     - Tenant Definition: Within the context of Dify, one tenant corresponds to one workspace. The workspace provides a separated area for each tenant's data and configurations.
-
-b. LOGO and copyright information: In the process of using Dify's frontend components, you may not remove or modify the LOGO or copyright information in the Dify console or applications. This restriction is inapplicable to uses of Dify that do not involve its frontend components.
+    
+b. LOGO and copyright information: In the process of using Dify's frontend, you may not remove or modify the LOGO or copyright information in the Dify console or applications. This restriction is inapplicable to uses of Dify that do not involve its frontend.
+    - Frontend Definition: For the purposes of this license, the "frontend" of Dify includes all components located in the `web/` directory when running Dify from the raw source code, or the "web" image when running Dify with Docker.
 
 Please contact business@dify.ai by email to inquire about licensing matters.
 

api/.env.example

@@ -42,7 +42,7 @@ DB_DATABASE=dify
 
 # Storage configuration
 # use for store upload files, private keys...
-# storage type: local, s3, azure-blob, google-storage, tencent-cos, huawei-obs, volcengine-tos, baidu-obs, supabase
+# storage type: local, s3, aliyun-oss, azure-blob, baidu-obs, google-storage, huawei-obs, oci-storage, tencent-cos, volcengine-tos, supabase
 STORAGE_TYPE=local
 STORAGE_LOCAL_PATH=storage
 S3_USE_AWS_MANAGED_IAM=false

api/README.md

@@ -85,3 +85,4 @@
    cd ../
    poetry run -C api bash dev/pytest/pytest_all_tests.sh
    ```
+

api/app.py

@@ -10,44 +10,20 @@ if os.environ.get("DEBUG", "false").lower() != "true":
     grpc.experimental.gevent.init_gevent()
 
 import json
-import logging
-import sys
 import threading
 import time
 import warnings
-from logging.handlers import RotatingFileHandler
 
-from flask import Flask, Response, request
-from flask_cors import CORS
-from werkzeug.exceptions import Unauthorized
+from flask import Response
 
-import contexts
-from commands import register_commands
-from configs import dify_config
+from app_factory import create_app
 
 # DO NOT REMOVE BELOW
 from events import event_handlers  # noqa: F401
-from extensions import (
-    ext_celery,
-    ext_code_based_extension,
-    ext_compress,
-    ext_database,
-    ext_hosting_provider,
-    ext_login,
-    ext_mail,
-    ext_migrate,
-    ext_proxy_fix,
-    ext_redis,
-    ext_sentry,
-    ext_storage,
-)
 from extensions.ext_database import db
-from extensions.ext_login import login_manager
-from libs.passport import PassportService
 
 # TODO: Find a way to avoid importing models here
 from models import account, dataset, model, source, task, tool, tools, web  # noqa: F401
-from services.account_service import AccountService
 
 # DO NOT REMOVE ABOVE
 
@@ -60,188 +36,12 @@ if hasattr(time, "tzset"):
     time.tzset()
 
 
-class DifyApp(Flask):
-    pass
-
-
 # -------------
 # Configuration
 # -------------
-
-
 config_type = os.getenv("EDITION", default="SELF_HOSTED")  # ce edition first
 
 
-# ----------------------------
-# Application Factory Function
-# ----------------------------
-
-
-def create_flask_app_with_configs() -> Flask:
-    """
-    create a raw flask app
-    with configs loaded from .env file
-    """
-    dify_app = DifyApp(__name__)
-    dify_app.config.from_mapping(dify_config.model_dump())
-
-    # populate configs into system environment variables
-    for key, value in dify_app.config.items():
-        if isinstance(value, str):
-            os.environ[key] = value
-        elif isinstance(value, int | float | bool):
-            os.environ[key] = str(value)
-        elif value is None:
-            os.environ[key] = ""
-
-    return dify_app
-
-
-def create_app() -> Flask:
-    app = create_flask_app_with_configs()
-
-    app.secret_key = app.config["SECRET_KEY"]
-
-    log_handlers = None
-    log_file = app.config.get("LOG_FILE")
-    if log_file:
-        log_dir = os.path.dirname(log_file)
-        os.makedirs(log_dir, exist_ok=True)
-        log_handlers = [
-            RotatingFileHandler(
-                filename=log_file,
-                maxBytes=1024 * 1024 * 1024,
-                backupCount=5,
-            ),
-            logging.StreamHandler(sys.stdout),
-        ]
-
-    logging.basicConfig(
-        level=app.config.get("LOG_LEVEL"),
-        format=app.config.get("LOG_FORMAT"),
-        datefmt=app.config.get("LOG_DATEFORMAT"),
-        handlers=log_handlers,
-        force=True,
-    )
-    log_tz = app.config.get("LOG_TZ")
-    if log_tz:
-        from datetime import datetime
-
-        import pytz
-
-        timezone = pytz.timezone(log_tz)
-
-        def time_converter(seconds):
-            return datetime.utcfromtimestamp(seconds).astimezone(timezone).timetuple()
-
-        for handler in logging.root.handlers:
-            handler.formatter.converter = time_converter
-    initialize_extensions(app)
-    register_blueprints(app)
-    register_commands(app)
-
-    return app
-
-
-def initialize_extensions(app):
-    # Since the application instance is now created, pass it to each Flask
-    # extension instance to bind it to the Flask application instance (app)
-    ext_compress.init_app(app)
-    ext_code_based_extension.init()
-    ext_database.init_app(app)
-    ext_migrate.init(app, db)
-    ext_redis.init_app(app)
-    ext_storage.init_app(app)
-    ext_celery.init_app(app)
-    ext_login.init_app(app)
-    ext_mail.init_app(app)
-    ext_hosting_provider.init_app(app)
-    ext_sentry.init_app(app)
-    ext_proxy_fix.init_app(app)
-
-
-# Flask-Login configuration
-@login_manager.request_loader
-def load_user_from_request(request_from_flask_login):
-    """Load user based on the request."""
-    if request.blueprint not in {"console", "inner_api"}:
-        return None
-    # Check if the user_id contains a dot, indicating the old format
-    auth_header = request.headers.get("Authorization", "")
-    if not auth_header:
-        auth_token = request.args.get("_token")
-        if not auth_token:
-            raise Unauthorized("Invalid Authorization token.")
-    else:
-        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.")
-
-    decoded = PassportService().verify(auth_token)
-    user_id = decoded.get("user_id")
-
-    logged_in_account = AccountService.load_logged_in_account(account_id=user_id)
-    if logged_in_account:
-        contexts.tenant_id.set(logged_in_account.current_tenant_id)
-    return logged_in_account
-
-
-@login_manager.unauthorized_handler
-def unauthorized_handler():
-    """Handle unauthorized requests."""
-    return Response(
-        json.dumps({"code": "unauthorized", "message": "Unauthorized."}),
-        status=401,
-        content_type="application/json",
-    )
-
-
-# register blueprint routers
-def register_blueprints(app):
-    from controllers.console import bp as console_app_bp
-    from controllers.files import bp as files_bp
-    from controllers.inner_api import bp as inner_api_bp
-    from controllers.service_api import bp as service_api_bp
-    from controllers.web import bp as web_bp
-
-    CORS(
-        service_api_bp,
-        allow_headers=["Content-Type", "Authorization", "X-App-Code"],
-        methods=["GET", "PUT", "POST", "DELETE", "OPTIONS", "PATCH"],
-    )
-    app.register_blueprint(service_api_bp)
-
-    CORS(
-        web_bp,
-        resources={r"/*": {"origins": app.config["WEB_API_CORS_ALLOW_ORIGINS"]}},
-        supports_credentials=True,
-        allow_headers=["Content-Type", "Authorization", "X-App-Code"],
-        methods=["GET", "PUT", "POST", "DELETE", "OPTIONS", "PATCH"],
-        expose_headers=["X-Version", "X-Env"],
-    )
-
-    app.register_blueprint(web_bp)
-
-    CORS(
-        console_app_bp,
-        resources={r"/*": {"origins": app.config["CONSOLE_CORS_ALLOW_ORIGINS"]}},
-        supports_credentials=True,
-        allow_headers=["Content-Type", "Authorization"],
-        methods=["GET", "PUT", "POST", "DELETE", "OPTIONS", "PATCH"],
-        expose_headers=["X-Version", "X-Env"],
-    )
-
-    app.register_blueprint(console_app_bp)
-
-    CORS(files_bp, allow_headers=["Content-Type"], methods=["GET", "PUT", "POST", "DELETE", "OPTIONS", "PATCH"])
-    app.register_blueprint(files_bp)
-
-    app.register_blueprint(inner_api_bp)
-
-
 # create app
 app = create_app()
 celery = app.extensions["celery"]

api/app_factory.py

@@ -0,0 +1,213 @@
+import os
+
+if os.environ.get("DEBUG", "false").lower() != "true":
+    from gevent import monkey
+
+    monkey.patch_all()
+
+    import grpc.experimental.gevent
+
+    grpc.experimental.gevent.init_gevent()
+
+import json
+import logging
+import sys
+from logging.handlers import RotatingFileHandler
+
+from flask import Flask, Response, request
+from flask_cors import CORS
+from werkzeug.exceptions import Unauthorized
+
+import contexts
+from commands import register_commands
+from configs import dify_config
+from extensions import (
+    ext_celery,
+    ext_code_based_extension,
+    ext_compress,
+    ext_database,
+    ext_hosting_provider,
+    ext_login,
+    ext_mail,
+    ext_migrate,
+    ext_proxy_fix,
+    ext_redis,
+    ext_sentry,
+    ext_storage,
+)
+from extensions.ext_database import db
+from extensions.ext_login import login_manager
+from libs.passport import PassportService
+from services.account_service import AccountService
+
+
+class DifyApp(Flask):
+    pass
+
+
+# ----------------------------
+# Application Factory Function
+# ----------------------------
+def create_flask_app_with_configs() -> Flask:
+    """
+    create a raw flask app
+    with configs loaded from .env file
+    """
+    dify_app = DifyApp(__name__)
+    dify_app.config.from_mapping(dify_config.model_dump())
+
+    # populate configs into system environment variables
+    for key, value in dify_app.config.items():
+        if isinstance(value, str):
+            os.environ[key] = value
+        elif isinstance(value, int | float | bool):
+            os.environ[key] = str(value)
+        elif value is None:
+            os.environ[key] = ""
+
+    return dify_app
+
+
+def create_app() -> Flask:
+    app = create_flask_app_with_configs()
+
+    app.secret_key = app.config["SECRET_KEY"]
+
+    log_handlers = None
+    log_file = app.config.get("LOG_FILE")
+    if log_file:
+        log_dir = os.path.dirname(log_file)
+        os.makedirs(log_dir, exist_ok=True)
+        log_handlers = [
+            RotatingFileHandler(
+                filename=log_file,
+                maxBytes=1024 * 1024 * 1024,
+                backupCount=5,
+            ),
+            logging.StreamHandler(sys.stdout),
+        ]
+
+    logging.basicConfig(
+        level=app.config.get("LOG_LEVEL"),
+        format=app.config.get("LOG_FORMAT"),
+        datefmt=app.config.get("LOG_DATEFORMAT"),
+        handlers=log_handlers,
+        force=True,
+    )
+    log_tz = app.config.get("LOG_TZ")
+    if log_tz:
+        from datetime import datetime
+
+        import pytz
+
+        timezone = pytz.timezone(log_tz)
+
+        def time_converter(seconds):
+            return datetime.utcfromtimestamp(seconds).astimezone(timezone).timetuple()
+
+        for handler in logging.root.handlers:
+            handler.formatter.converter = time_converter
+    initialize_extensions(app)
+    register_blueprints(app)
+    register_commands(app)
+
+    return app
+
+
+def initialize_extensions(app):
+    # Since the application instance is now created, pass it to each Flask
+    # extension instance to bind it to the Flask application instance (app)
+    ext_compress.init_app(app)
+    ext_code_based_extension.init()
+    ext_database.init_app(app)
+    ext_migrate.init(app, db)
+    ext_redis.init_app(app)
+    ext_storage.init_app(app)
+    ext_celery.init_app(app)
+    ext_login.init_app(app)
+    ext_mail.init_app(app)
+    ext_hosting_provider.init_app(app)
+    ext_sentry.init_app(app)
+    ext_proxy_fix.init_app(app)
+
+
+# Flask-Login configuration
+@login_manager.request_loader
+def load_user_from_request(request_from_flask_login):
+    """Load user based on the request."""
+    if request.blueprint not in {"console", "inner_api"}:
+        return None
+    # Check if the user_id contains a dot, indicating the old format
+    auth_header = request.headers.get("Authorization", "")
+    if not auth_header:
+        auth_token = request.args.get("_token")
+        if not auth_token:
+            raise Unauthorized("Invalid Authorization token.")
+    else:
+        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.")
+
+    decoded = PassportService().verify(auth_token)
+    user_id = decoded.get("user_id")
+
+    logged_in_account = AccountService.load_logged_in_account(account_id=user_id)
+    if logged_in_account:
+        contexts.tenant_id.set(logged_in_account.current_tenant_id)
+    return logged_in_account
+
+
+@login_manager.unauthorized_handler
+def unauthorized_handler():
+    """Handle unauthorized requests."""
+    return Response(
+        json.dumps({"code": "unauthorized", "message": "Unauthorized."}),
+        status=401,
+        content_type="application/json",
+    )
+
+
+# register blueprint routers
+def register_blueprints(app):
+    from controllers.console import bp as console_app_bp
+    from controllers.files import bp as files_bp
+    from controllers.inner_api import bp as inner_api_bp
+    from controllers.service_api import bp as service_api_bp
+    from controllers.web import bp as web_bp
+
+    CORS(
+        service_api_bp,
+        allow_headers=["Content-Type", "Authorization", "X-App-Code"],
+        methods=["GET", "PUT", "POST", "DELETE", "OPTIONS", "PATCH"],
+    )
+    app.register_blueprint(service_api_bp)
+
+    CORS(
+        web_bp,
+        resources={r"/*": {"origins": app.config["WEB_API_CORS_ALLOW_ORIGINS"]}},
+        supports_credentials=True,
+        allow_headers=["Content-Type", "Authorization", "X-App-Code"],
+        methods=["GET", "PUT", "POST", "DELETE", "OPTIONS", "PATCH"],
+        expose_headers=["X-Version", "X-Env"],
+    )
+
+    app.register_blueprint(web_bp)
+
+    CORS(
+        console_app_bp,
+        resources={r"/*": {"origins": app.config["CONSOLE_CORS_ALLOW_ORIGINS"]}},
+        supports_credentials=True,
+        allow_headers=["Content-Type", "Authorization"],
+        methods=["GET", "PUT", "POST", "DELETE", "OPTIONS", "PATCH"],
+        expose_headers=["X-Version", "X-Env"],
+    )
+
+    app.register_blueprint(console_app_bp)
+
+    CORS(files_bp, allow_headers=["Content-Type"], methods=["GET", "PUT", "POST", "DELETE", "OPTIONS", "PATCH"])
+    app.register_blueprint(files_bp)
+
+    app.register_blueprint(inner_api_bp)

api/commands.py

@@ -259,6 +259,25 @@ def migrate_knowledge_vector_database():
     skipped_count = 0
     total_count = 0
     vector_type = dify_config.VECTOR_STORE
+    upper_colletion_vector_types = {
+        VectorType.MILVUS,
+        VectorType.PGVECTOR,
+        VectorType.RELYT,
+        VectorType.WEAVIATE,
+        VectorType.ORACLE,
+        VectorType.ELASTICSEARCH,
+    }
+    lower_colletion_vector_types = {
+        VectorType.ANALYTICDB,
+        VectorType.CHROMA,
+        VectorType.MYSCALE,
+        VectorType.PGVECTO_RS,
+        VectorType.TIDB_VECTOR,
+        VectorType.OPENSEARCH,
+        VectorType.TENCENT,
+        VectorType.BAIDU,
+        VectorType.VIKINGDB,
+    }
     page = 1
     while True:
         try:
@@ -284,11 +303,9 @@ def migrate_knowledge_vector_database():
                         skipped_count = skipped_count + 1
                         continue
                 collection_name = ""
-                if vector_type == VectorType.WEAVIATE:
-                    dataset_id = dataset.id
+                dataset_id = dataset.id
+                if vector_type in upper_colletion_vector_types:
                     collection_name = Dataset.gen_collection_name_by_id(dataset_id)
-                    index_struct_dict = {"type": VectorType.WEAVIATE, "vector_store": {"class_prefix": collection_name}}
-                    dataset.index_struct = json.dumps(index_struct_dict)
                 elif vector_type == VectorType.QDRANT:
                     if dataset.collection_binding_id:
                         dataset_collection_binding = (
@@ -301,63 +318,15 @@ def migrate_knowledge_vector_database():
                         else:
                             raise ValueError("Dataset Collection Binding not found")
                     else:
-                        dataset_id = dataset.id
                         collection_name = Dataset.gen_collection_name_by_id(dataset_id)
-                    index_struct_dict = {"type": VectorType.QDRANT, "vector_store": {"class_prefix": collection_name}}
-                    dataset.index_struct = json.dumps(index_struct_dict)
 
-                elif vector_type == VectorType.MILVUS:
-                    dataset_id = dataset.id
-                    collection_name = Dataset.gen_collection_name_by_id(dataset_id)
-                    index_struct_dict = {"type": VectorType.MILVUS, "vector_store": {"class_prefix": collection_name}}
-                    dataset.index_struct = json.dumps(index_struct_dict)
-                elif vector_type == VectorType.RELYT:
-                    dataset_id = dataset.id
-                    collection_name = Dataset.gen_collection_name_by_id(dataset_id)
-                    index_struct_dict = {"type": "relyt", "vector_store": {"class_prefix": collection_name}}
-                    dataset.index_struct = json.dumps(index_struct_dict)
-                elif vector_type == VectorType.TENCENT:
-                    dataset_id = dataset.id
-                    collection_name = Dataset.gen_collection_name_by_id(dataset_id)
-                    index_struct_dict = {"type": VectorType.TENCENT, "vector_store": {"class_prefix": collection_name}}
-                    dataset.index_struct = json.dumps(index_struct_dict)
-                elif vector_type == VectorType.PGVECTOR:
-                    dataset_id = dataset.id
-                    collection_name = Dataset.gen_collection_name_by_id(dataset_id)
-                    index_struct_dict = {"type": VectorType.PGVECTOR, "vector_store": {"class_prefix": collection_name}}
-                    dataset.index_struct = json.dumps(index_struct_dict)
-                elif vector_type == VectorType.OPENSEARCH:
-                    dataset_id = dataset.id
-                    collection_name = Dataset.gen_collection_name_by_id(dataset_id)
-                    index_struct_dict = {
-                        "type": VectorType.OPENSEARCH,
-                        "vector_store": {"class_prefix": collection_name},
-                    }
-                    dataset.index_struct = json.dumps(index_struct_dict)
-                elif vector_type == VectorType.ANALYTICDB:
-                    dataset_id = dataset.id
-                    collection_name = Dataset.gen_collection_name_by_id(dataset_id)
-                    index_struct_dict = {
-                        "type": VectorType.ANALYTICDB,
-                        "vector_store": {"class_prefix": collection_name},
-                    }
-                    dataset.index_struct = json.dumps(index_struct_dict)
-                elif vector_type == VectorType.ELASTICSEARCH:
-                    dataset_id = dataset.id
-                    index_name = Dataset.gen_collection_name_by_id(dataset_id)
-                    index_struct_dict = {"type": "elasticsearch", "vector_store": {"class_prefix": index_name}}
-                    dataset.index_struct = json.dumps(index_struct_dict)
-                elif vector_type == VectorType.BAIDU:
-                    dataset_id = dataset.id
-                    collection_name = Dataset.gen_collection_name_by_id(dataset_id)
-                    index_struct_dict = {
-                        "type": VectorType.BAIDU,
-                        "vector_store": {"class_prefix": collection_name},
-                    }
-                    dataset.index_struct = json.dumps(index_struct_dict)
+                elif vector_type in lower_colletion_vector_types:
+                    collection_name = Dataset.gen_collection_name_by_id(dataset_id).lower()
                 else:
                     raise ValueError(f"Vector store {vector_type} is not supported.")
 
+                index_struct_dict = {"type": vector_type, "vector_store": {"class_prefix": collection_name}}
+                dataset.index_struct = json.dumps(index_struct_dict)
                 vector = Vector(dataset)
                 click.echo(f"Migrating dataset {dataset.id}.")
 

api/configs/feature/__init__.py

@@ -506,11 +506,16 @@ class DataSetConfig(BaseSettings):
     Configuration for dataset management
     """
 
-    CLEAN_DAY_SETTING: PositiveInt = Field(
-        description="Interval in days for dataset cleanup operations",
+    PLAN_SANDBOX_CLEAN_DAY_SETTING: PositiveInt = Field(
+        description="Interval in days for dataset cleanup operations - plan: sandbox",
         default=30,
     )
 
+    PLAN_PRO_CLEAN_DAY_SETTING: PositiveInt = Field(
+        description="Interval in days for dataset cleanup operations - plan: pro and team",
+        default=7,
+    )
+
     DATASET_OPERATOR_ENABLED: bool = Field(
         description="Enable or disable dataset operator functionality",
         default=False,

api/configs/middleware/__init__.py

@@ -35,7 +35,8 @@ from configs.middleware.vdb.weaviate_config import WeaviateConfig
 class StorageConfig(BaseSettings):
     STORAGE_TYPE: str = Field(
         description="Type of storage to use."
-        " Options: 'local', 's3', 'azure-blob', 'aliyun-oss', 'google-storage'. Default is 'local'.",
+        " Options: 'local', 's3', 'aliyun-oss', 'azure-blob', 'baidu-obs', 'google-storage', 'huawei-obs', "
+        "'oci-storage', 'tencent-cos', 'volcengine-tos', 'supabase'. Default is 'local'.",
         default="local",
     )
 

api/configs/middleware/vdb/oracle_config.py

@@ -14,7 +14,7 @@ class OracleConfig(BaseSettings):
         default=None,
     )
 
-    ORACLE_PORT: Optional[PositiveInt] = Field(
+    ORACLE_PORT: PositiveInt = Field(
         description="Port number on which the Oracle database server is listening (default is 1521)",
         default=1521,
     )

api/configs/middleware/vdb/pgvector_config.py

@@ -14,7 +14,7 @@ class PGVectorConfig(BaseSettings):
         default=None,
     )
 
-    PGVECTOR_PORT: Optional[PositiveInt] = Field(
+    PGVECTOR_PORT: PositiveInt = Field(
         description="Port number on which the PostgreSQL server is listening (default is 5433)",
         default=5433,
     )

api/configs/middleware/vdb/pgvectors_config.py

@@ -14,7 +14,7 @@ class PGVectoRSConfig(BaseSettings):
         default=None,
     )
 
-    PGVECTO_RS_PORT: Optional[PositiveInt] = Field(
+    PGVECTO_RS_PORT: PositiveInt = Field(
         description="Port number on which the PostgreSQL server with PGVecto.RS is listening (default is 5431)",
         default=5431,
     )

api/configs/middleware/vdb/vikingdb_config.py

@@ -11,27 +11,39 @@ class VikingDBConfig(BaseModel):
     """
 
     VIKINGDB_ACCESS_KEY: Optional[str] = Field(
-        default=None, description="The Access Key provided by Volcengine VikingDB for API authentication."
+        description="The Access Key provided by Volcengine VikingDB for API authentication."
+        "Refer to the following documentation for details on obtaining credentials:"
+        "https://www.volcengine.com/docs/6291/65568",
+        default=None,
     )
+
     VIKINGDB_SECRET_KEY: Optional[str] = Field(
-        default=None, description="The Secret Key provided by Volcengine VikingDB for API authentication."
+        description="The Secret Key provided by Volcengine VikingDB for API authentication.",
+        default=None,
     )
-    VIKINGDB_REGION: Optional[str] = Field(
-        default="cn-shanghai",
+
+    VIKINGDB_REGION: str = Field(
         description="The region of the Volcengine VikingDB service.(e.g., 'cn-shanghai', 'cn-beijing').",
+        default="cn-shanghai",
     )
-    VIKINGDB_HOST: Optional[str] = Field(
-        default="api-vikingdb.mlp.cn-shanghai.volces.com",
+
+    VIKINGDB_HOST: str = Field(
         description="The host of the Volcengine VikingDB service.(e.g., 'api-vikingdb.volces.com', \
             'api-vikingdb.mlp.cn-shanghai.volces.com')",
+        default="api-vikingdb.mlp.cn-shanghai.volces.com",
     )
-    VIKINGDB_SCHEME: Optional[str] = Field(
-        default="http",
+
+    VIKINGDB_SCHEME: str = Field(
         description="The scheme of the Volcengine VikingDB service.(e.g., 'http', 'https').",
+        default="http",
     )
-    VIKINGDB_CONNECTION_TIMEOUT: Optional[int] = Field(
-        default=30, description="The connection timeout of the Volcengine VikingDB service."
+
+    VIKINGDB_CONNECTION_TIMEOUT: int = Field(
+        description="The connection timeout of the Volcengine VikingDB service.",
+        default=30,
     )
-    VIKINGDB_SOCKET_TIMEOUT: Optional[int] = Field(
-        default=30, description="The socket timeout of the Volcengine VikingDB service."
+
+    VIKINGDB_SOCKET_TIMEOUT: int = Field(
+        description="The socket timeout of the Volcengine VikingDB service.",
+        default=30,
     )

api/controllers/console/datasets/hit_testing.py

@@ -1,88 +1,24 @@
-import logging
+from flask_restful import Resource
 
-from flask_login import current_user
-from flask_restful import Resource, marshal, reqparse
-from werkzeug.exceptions import Forbidden, InternalServerError, NotFound
-
-import services
 from controllers.console import api
-from controllers.console.app.error import (
-    CompletionRequestError,
-    ProviderModelCurrentlyNotSupportError,
-    ProviderNotInitializeError,
-    ProviderQuotaExceededError,
-)
-from controllers.console.datasets.error import DatasetNotInitializedError
+from controllers.console.datasets.hit_testing_base import DatasetsHitTestingBase
 from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required
-from core.errors.error import (
-    LLMBadRequestError,
-    ModelCurrentlyNotSupportError,
-    ProviderTokenNotInitError,
-    QuotaExceededError,
-)
-from core.model_runtime.errors.invoke import InvokeError
-from fields.hit_testing_fields import hit_testing_record_fields
 from libs.login import login_required
-from services.dataset_service import DatasetService
-from services.hit_testing_service import HitTestingService
 
 
-class HitTestingApi(Resource):
+class HitTestingApi(Resource, DatasetsHitTestingBase):
     @setup_required
     @login_required
     @account_initialization_required
     def post(self, dataset_id):
         dataset_id_str = str(dataset_id)
 
-        dataset = DatasetService.get_dataset(dataset_id_str)
-        if dataset is None:
-            raise NotFound("Dataset not found.")
+        dataset = self.get_and_validate_dataset(dataset_id_str)
+        args = self.parse_args()
+        self.hit_testing_args_check(args)
 
-        try:
-            DatasetService.check_dataset_permission(dataset, current_user)
-        except services.errors.account.NoPermissionError as e:
-            raise Forbidden(str(e))
-
-        parser = reqparse.RequestParser()
-        parser.add_argument("query", type=str, location="json")
-        parser.add_argument("retrieval_model", type=dict, required=False, location="json")
-        parser.add_argument("external_retrieval_model", type=dict, required=False, location="json")
-        args = parser.parse_args()
-
-        HitTestingService.hit_testing_args_check(args)
-
-        try:
-            response = HitTestingService.retrieve(
-                dataset=dataset,
-                query=args["query"],
-                account=current_user,
-                retrieval_model=args["retrieval_model"],
-                external_retrieval_model=args["external_retrieval_model"],
-                limit=10,
-            )
-
-            return {"query": response["query"], "records": marshal(response["records"], hit_testing_record_fields)}
-        except services.errors.index.IndexNotInitializedError:
-            raise DatasetNotInitializedError()
-        except ProviderTokenNotInitError as ex:
-            raise ProviderNotInitializeError(ex.description)
-        except QuotaExceededError:
-            raise ProviderQuotaExceededError()
-        except ModelCurrentlyNotSupportError:
-            raise ProviderModelCurrentlyNotSupportError()
-        except LLMBadRequestError:
-            raise ProviderNotInitializeError(
-                "No Embedding Model or Reranking Model available. Please configure a valid provider "
-                "in the Settings -> Model Provider."
-            )
-        except InvokeError as e:
-            raise CompletionRequestError(e.description)
-        except ValueError as e:
-            raise ValueError(str(e))
-        except Exception as e:
-            logging.exception("Hit testing failed.")
-            raise InternalServerError(str(e))
+        return self.perform_hit_testing(dataset, args)
 
 
 api.add_resource(HitTestingApi, "/datasets/<uuid:dataset_id>/hit-testing")

api/controllers/console/datasets/hit_testing_base.py

@@ -0,0 +1,85 @@
+import logging
+
+from flask_login import current_user
+from flask_restful import marshal, reqparse
+from werkzeug.exceptions import Forbidden, InternalServerError, NotFound
+
+import services.dataset_service
+from controllers.console.app.error import (
+    CompletionRequestError,
+    ProviderModelCurrentlyNotSupportError,
+    ProviderNotInitializeError,
+    ProviderQuotaExceededError,
+)
+from controllers.console.datasets.error import DatasetNotInitializedError
+from core.errors.error import (
+    LLMBadRequestError,
+    ModelCurrentlyNotSupportError,
+    ProviderTokenNotInitError,
+    QuotaExceededError,
+)
+from core.model_runtime.errors.invoke import InvokeError
+from fields.hit_testing_fields import hit_testing_record_fields
+from services.dataset_service import DatasetService
+from services.hit_testing_service import HitTestingService
+
+
+class DatasetsHitTestingBase:
+    @staticmethod
+    def get_and_validate_dataset(dataset_id: str):
+        dataset = DatasetService.get_dataset(dataset_id)
+        if dataset is None:
+            raise NotFound("Dataset not found.")
+
+        try:
+            DatasetService.check_dataset_permission(dataset, current_user)
+        except services.errors.account.NoPermissionError as e:
+            raise Forbidden(str(e))
+
+        return dataset
+
+    @staticmethod
+    def hit_testing_args_check(args):
+        HitTestingService.hit_testing_args_check(args)
+
+    @staticmethod
+    def parse_args():
+        parser = reqparse.RequestParser()
+
+        parser.add_argument("query", type=str, location="json")
+        parser.add_argument("retrieval_model", type=dict, required=False, location="json")
+        parser.add_argument("external_retrieval_model", type=dict, required=False, location="json")
+        return parser.parse_args()
+
+    @staticmethod
+    def perform_hit_testing(dataset, args):
+        try:
+            response = HitTestingService.retrieve(
+                dataset=dataset,
+                query=args["query"],
+                account=current_user,
+                retrieval_model=args["retrieval_model"],
+                external_retrieval_model=args["external_retrieval_model"],
+                limit=10,
+            )
+            return {"query": response["query"], "records": marshal(response["records"], hit_testing_record_fields)}
+        except services.errors.index.IndexNotInitializedError:
+            raise DatasetNotInitializedError()
+        except ProviderTokenNotInitError as ex:
+            raise ProviderNotInitializeError(ex.description)
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
+        except LLMBadRequestError:
+            raise ProviderNotInitializeError(
+                "No Embedding Model or Reranking Model available. Please configure a valid provider "
+                "in the Settings -> Model Provider."
+            )
+        except InvokeError as e:
+            raise CompletionRequestError(e.description)
+        except ValueError as e:
+            raise ValueError(str(e))
+        except Exception as e:
+            logging.exception("Hit testing failed.")
+            raise InternalServerError(str(e))

api/controllers/service_api/__init__.py

@@ -5,7 +5,6 @@ from libs.external_api import ExternalApi
 bp = Blueprint("service_api", __name__, url_prefix="/v1")
 api = ExternalApi(bp)
 
-
 from . import index
 from .app import app, audio, completion, conversation, file, message, workflow
-from .dataset import dataset, document, segment
+from .dataset import dataset, document, hit_testing, segment

api/controllers/service_api/app/completion.py

@@ -4,7 +4,6 @@ from flask_restful import Resource, reqparse
 from werkzeug.exceptions import InternalServerError, NotFound
 
 import services
-from constants import UUID_NIL
 from controllers.service_api import api
 from controllers.service_api.app.error import (
     AppUnavailableError,
@@ -108,7 +107,6 @@ class ChatApi(Resource):
         parser.add_argument("conversation_id", type=uuid_value, location="json")
         parser.add_argument("retriever_from", type=str, required=False, default="dev", location="json")
         parser.add_argument("auto_generate_name", type=bool, required=False, default=True, location="json")
-        parser.add_argument("parent_message_id", type=uuid_value, required=False, default=UUID_NIL, location="json")
 
         args = parser.parse_args()
 

api/controllers/service_api/dataset/hit_testing.py

@@ -0,0 +1,17 @@
+from controllers.console.datasets.hit_testing_base import DatasetsHitTestingBase
+from controllers.service_api import api
+from controllers.service_api.wraps import DatasetApiResource
+
+
+class HitTestingApi(DatasetApiResource, DatasetsHitTestingBase):
+    def post(self, tenant_id, dataset_id):
+        dataset_id_str = str(dataset_id)
+
+        dataset = self.get_and_validate_dataset(dataset_id_str)
+        args = self.parse_args()
+        self.hit_testing_args_check(args)
+
+        return self.perform_hit_testing(dataset, args)
+
+
+api.add_resource(HitTestingApi, "/datasets/<uuid:dataset_id>/hit-testing")

api/core/agent/output_parser/cot_output_parser.py

@@ -62,6 +62,8 @@ class CotAgentOutputParser:
         thought_str = "thought:"
         thought_idx = 0
 
+        last_character = ""
+
         for response in llm_response:
             if response.delta.usage:
                 usage_dict["usage"] = response.delta.usage
@@ -74,35 +76,38 @@ class CotAgentOutputParser:
             while index < len(response):
                 steps = 1
                 delta = response[index : index + steps]
-                last_character = response[index - 1] if index > 0 else ""
+                yield_delta = False
 
                 if delta == "`":
+                    last_character = delta
                     code_block_cache += delta
                     code_block_delimiter_count += 1
                 else:
                     if not in_code_block:
                         if code_block_delimiter_count > 0:
+                            last_character = delta
                             yield code_block_cache
                         code_block_cache = ""
                     else:
+                        last_character = delta
                         code_block_cache += delta
                     code_block_delimiter_count = 0
 
                 if not in_code_block and not in_json:
                     if delta.lower() == action_str[action_idx] and action_idx == 0:
                         if last_character not in {"\n", " ", ""}:
+                            yield_delta = True
+                        else:
+                            last_character = delta
+                            action_cache += delta
+                            action_idx += 1
+                            if action_idx == len(action_str):
+                                action_cache = ""
+                                action_idx = 0
                             index += steps
-                            yield delta
                             continue
-
-                        action_cache += delta
-                        action_idx += 1
-                        if action_idx == len(action_str):
-                            action_cache = ""
-                            action_idx = 0
-                        index += steps
-                        continue
                     elif delta.lower() == action_str[action_idx] and action_idx > 0:
+                        last_character = delta
                         action_cache += delta
                         action_idx += 1
                         if action_idx == len(action_str):
@@ -112,24 +117,25 @@ class CotAgentOutputParser:
                         continue
                     else:
                         if action_cache:
+                            last_character = delta
                             yield action_cache
                             action_cache = ""
                             action_idx = 0
 
                     if delta.lower() == thought_str[thought_idx] and thought_idx == 0:
                         if last_character not in {"\n", " ", ""}:
+                            yield_delta = True
+                        else:
+                            last_character = delta
+                            thought_cache += delta
+                            thought_idx += 1
+                            if thought_idx == len(thought_str):
+                                thought_cache = ""
+                                thought_idx = 0
                             index += steps
-                            yield delta
                             continue
-
-                        thought_cache += delta
-                        thought_idx += 1
-                        if thought_idx == len(thought_str):
-                            thought_cache = ""
-                            thought_idx = 0
-                        index += steps
-                        continue
                     elif delta.lower() == thought_str[thought_idx] and thought_idx > 0:
+                        last_character = delta
                         thought_cache += delta
                         thought_idx += 1
                         if thought_idx == len(thought_str):
@@ -139,12 +145,20 @@ class CotAgentOutputParser:
                         continue
                     else:
                         if thought_cache:
+                            last_character = delta
                             yield thought_cache
                             thought_cache = ""
                             thought_idx = 0
 
+                    if yield_delta:
+                        index += steps
+                        last_character = delta
+                        yield delta
+                        continue
+
                 if code_block_delimiter_count == 3:
                     if in_code_block:
+                        last_character = delta
                         yield from extra_json_from_code_block(code_block_cache)
                         code_block_cache = ""
 
@@ -156,8 +170,10 @@ class CotAgentOutputParser:
                     if delta == "{":
                         json_quote_count += 1
                         in_json = True
+                        last_character = delta
                         json_cache += delta
                     elif delta == "}":
+                        last_character = delta
                         json_cache += delta
                         if json_quote_count > 0:
                             json_quote_count -= 1
@@ -168,16 +184,19 @@ class CotAgentOutputParser:
                                 continue
                     else:
                         if in_json:
+                            last_character = delta
                             json_cache += delta
 
                     if got_json:
                         got_json = False
+                        last_character = delta
                         yield parse_action(json_cache)
                         json_cache = ""
                         json_quote_count = 0
                         in_json = False
 
                 if not in_code_block and not in_json:
+                    last_character = delta
                     yield delta.replace("`", "")
 
                 index += steps

api/core/app/apps/advanced_chat/app_generator.py

@@ -10,6 +10,7 @@ from flask import Flask, current_app
 from pydantic import ValidationError
 
 import contexts
+from constants import UUID_NIL
 from core.app.app_config.features.file_upload.manager import FileUploadConfigManager
 from core.app.apps.advanced_chat.app_config_manager import AdvancedChatAppConfigManager
 from core.app.apps.advanced_chat.app_runner import AdvancedChatAppRunner
@@ -122,7 +123,7 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
             inputs=conversation.inputs if conversation else self._get_cleaned_inputs(inputs, app_config),
             query=query,
             files=file_objs,
-            parent_message_id=args.get("parent_message_id"),
+            parent_message_id=args.get("parent_message_id") if invoke_from != InvokeFrom.SERVICE_API else UUID_NIL,
             user_id=user.id,
             stream=stream,
             invoke_from=invoke_from,

api/core/app/apps/agent_chat/app_generator.py

@@ -8,6 +8,7 @@ from typing import Any, Literal, Union, overload
 from flask import Flask, current_app
 from pydantic import ValidationError
 
+from constants import UUID_NIL
 from core.app.app_config.easy_ui_based_app.model_config.converter import ModelConfigConverter
 from core.app.app_config.features.file_upload.manager import FileUploadConfigManager
 from core.app.apps.agent_chat.app_config_manager import AgentChatAppConfigManager
@@ -127,7 +128,7 @@ class AgentChatAppGenerator(MessageBasedAppGenerator):
             inputs=conversation.inputs if conversation else self._get_cleaned_inputs(inputs, app_config),
             query=query,
             files=file_objs,
-            parent_message_id=args.get("parent_message_id"),
+            parent_message_id=args.get("parent_message_id") if invoke_from != InvokeFrom.SERVICE_API else UUID_NIL,
             user_id=user.id,
             stream=stream,
             invoke_from=invoke_from,

api/core/app/apps/chat/app_generator.py

@@ -8,6 +8,7 @@ from typing import Any, Literal, Union, overload
 from flask import Flask, current_app
 from pydantic import ValidationError
 
+from constants import UUID_NIL
 from core.app.app_config.easy_ui_based_app.model_config.converter import ModelConfigConverter
 from core.app.app_config.features.file_upload.manager import FileUploadConfigManager
 from core.app.apps.base_app_queue_manager import AppQueueManager, GenerateTaskStoppedError, PublishFrom
@@ -128,7 +129,7 @@ class ChatAppGenerator(MessageBasedAppGenerator):
             inputs=conversation.inputs if conversation else self._get_cleaned_inputs(inputs, app_config),
             query=query,
             files=file_objs,
-            parent_message_id=args.get("parent_message_id"),
+            parent_message_id=args.get("parent_message_id") if invoke_from != InvokeFrom.SERVICE_API else UUID_NIL,
             user_id=user.id,
             stream=stream,
             invoke_from=invoke_from,

api/core/app/entities/app_invoke_entities.py

@@ -2,8 +2,9 @@ from collections.abc import Mapping
 from enum import Enum
 from typing import Any, Optional
 
-from pydantic import BaseModel, ConfigDict
+from pydantic import BaseModel, ConfigDict, Field, ValidationInfo, field_validator
 
+from constants import UUID_NIL
 from core.app.app_config.entities import AppConfig, EasyUIBasedAppConfig, WorkflowUIBasedAppConfig
 from core.entities.provider_configuration import ProviderModelBundle
 from core.file.file_obj import FileVar
@@ -116,13 +117,36 @@ class EasyUIBasedAppGenerateEntity(AppGenerateEntity):
     model_config = ConfigDict(protected_namespaces=())
 
 
-class ChatAppGenerateEntity(EasyUIBasedAppGenerateEntity):
+class ConversationAppGenerateEntity(AppGenerateEntity):
+    """
+    Base entity for conversation-based app generation.
+    """
+
+    conversation_id: Optional[str] = None
+    parent_message_id: Optional[str] = Field(
+        default=None,
+        description=(
+            "Starting from v0.9.0, parent_message_id is used to support message regeneration for internal chat API."
+            "For service API, we need to ensure its forward compatibility, "
+            "so passing in the parent_message_id as request arg is not supported for now. "
+            "It needs to be set to UUID_NIL so that the subsequent processing will treat it as legacy messages."
+        ),
+    )
+
+    @field_validator("parent_message_id")
+    @classmethod
+    def validate_parent_message_id(cls, v, info: ValidationInfo):
+        if info.data.get("invoke_from") == InvokeFrom.SERVICE_API and v != UUID_NIL:
+            raise ValueError("parent_message_id should be UUID_NIL for service API")
+        return v
+
+
+class ChatAppGenerateEntity(ConversationAppGenerateEntity, EasyUIBasedAppGenerateEntity):
     """
     Chat Application Generate Entity.
     """
 
-    conversation_id: Optional[str] = None
-    parent_message_id: Optional[str] = None
+    pass
 
 
 class CompletionAppGenerateEntity(EasyUIBasedAppGenerateEntity):
@@ -133,16 +157,15 @@ class CompletionAppGenerateEntity(EasyUIBasedAppGenerateEntity):
     pass
 
 
-class AgentChatAppGenerateEntity(EasyUIBasedAppGenerateEntity):
+class AgentChatAppGenerateEntity(ConversationAppGenerateEntity, EasyUIBasedAppGenerateEntity):
     """
     Agent Chat Application Generate Entity.
     """
 
-    conversation_id: Optional[str] = None
-    parent_message_id: Optional[str] = None
+    pass
 
 
-class AdvancedChatAppGenerateEntity(AppGenerateEntity):
+class AdvancedChatAppGenerateEntity(ConversationAppGenerateEntity):
     """
     Advanced Chat Application Generate Entity.
     """
@@ -150,8 +173,6 @@ class AdvancedChatAppGenerateEntity(AppGenerateEntity):
     # app config
     app_config: WorkflowUIBasedAppConfig
 
-    conversation_id: Optional[str] = None
-    parent_message_id: Optional[str] = None
     workflow_run_id: Optional[str] = None
     query: str
 

api/core/model_manager.py

@@ -3,7 +3,7 @@ import os
 from collections.abc import Callable, Generator, Sequence
 from typing import IO, Optional, Union, cast
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.entities.provider_configuration import ProviderConfiguration, ProviderModelBundle
 from core.entities.provider_entities import ModelLoadBalancingConfiguration
 from core.errors.error import ProviderTokenNotInitError

api/core/model_runtime/model_providers/__base/text_embedding_model.py

@@ -4,7 +4,7 @@ from typing import Optional
 
 from pydantic import ConfigDict
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.model_entities import ModelPropertyKey, ModelType
 from core.model_runtime.entities.text_embedding_entities import TextEmbeddingResult
 from core.model_runtime.model_providers.__base.ai_model import AIModel

api/core/model_runtime/model_providers/azure_openai/_constant.py

@@ -1098,6 +1098,14 @@ LLM_BASE_MODELS = [
                 ModelPropertyKey.CONTEXT_SIZE: 128000,
             },
             parameter_rules=[
+                ParameterRule(
+                    name="temperature",
+                    **PARAMETER_RULE_TEMPLATE[DefaultParameterName.TEMPERATURE],
+                ),
+                ParameterRule(
+                    name="top_p",
+                    **PARAMETER_RULE_TEMPLATE[DefaultParameterName.TOP_P],
+                ),
                 ParameterRule(
                     name="response_format",
                     label=I18nObject(zh_Hans="回复格式", en_US="response_format"),
@@ -1135,6 +1143,14 @@ LLM_BASE_MODELS = [
                 ModelPropertyKey.CONTEXT_SIZE: 128000,
             },
             parameter_rules=[
+                ParameterRule(
+                    name="temperature",
+                    **PARAMETER_RULE_TEMPLATE[DefaultParameterName.TEMPERATURE],
+                ),
+                ParameterRule(
+                    name="top_p",
+                    **PARAMETER_RULE_TEMPLATE[DefaultParameterName.TOP_P],
+                ),
                 ParameterRule(
                     name="response_format",
                     label=I18nObject(zh_Hans="回复格式", en_US="response_format"),

api/core/model_runtime/model_providers/azure_openai/llm/llm.py

@@ -119,7 +119,15 @@ class AzureOpenAILargeLanguageModel(_CommonAzureOpenAI, LargeLanguageModel):
         try:
             client = AzureOpenAI(**self._to_credential_kwargs(credentials))
 
-            if ai_model_entity.entity.model_properties.get(ModelPropertyKey.MODE) == LLMMode.CHAT.value:
+            if model.startswith("o1"):
+                client.chat.completions.create(
+                    messages=[{"role": "user", "content": "ping"}],
+                    model=model,
+                    temperature=1,
+                    max_completion_tokens=20,
+                    stream=False,
+                )
+            elif ai_model_entity.entity.model_properties.get(ModelPropertyKey.MODE) == LLMMode.CHAT.value:
                 # chat model
                 client.chat.completions.create(
                     messages=[{"role": "user", "content": "ping"}],

api/core/model_runtime/model_providers/azure_openai/text_embedding/text_embedding.py

@@ -7,7 +7,7 @@ import numpy as np
 import tiktoken
 from openai import AzureOpenAI
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.model_entities import AIModelEntity, PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult
 from core.model_runtime.errors.validate import CredentialsValidateFailedError

api/core/model_runtime/model_providers/baichuan/text_embedding/text_embedding.py

@@ -4,7 +4,7 @@ from typing import Optional
 
 from requests import post
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.model_entities import PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult
 from core.model_runtime.errors.invoke import (

api/core/model_runtime/model_providers/bedrock/llm/eu.anthropic.claude-3-haiku-v1.yaml

@@ -52,6 +52,8 @@ parameter_rules:
     help:
       zh_Hans: 对于每个后续标记，仅从前 K 个选项中进行采样。使用 top_k 删除长尾低概率响应。
       en_US: Only sample from the top K options for each subsequent token. Use top_k to remove long tail low probability responses.
+  - name: response_format
+    use_template: response_format
 pricing:
   input: '0.00025'
   output: '0.00125'

api/core/model_runtime/model_providers/bedrock/llm/eu.anthropic.claude-3-sonnet-v1.5.yaml

@@ -51,6 +51,8 @@ parameter_rules:
     help:
       zh_Hans: 对于每个后续标记，仅从前 K 个选项中进行采样。使用 top_k 删除长尾低概率响应。
       en_US: Only sample from the top K options for each subsequent token. Use top_k to remove long tail low probability responses.
+  - name: response_format
+    use_template: response_format
 pricing:
   input: '0.003'
   output: '0.015'

api/core/model_runtime/model_providers/bedrock/llm/eu.anthropic.claude-3-sonnet-v1.yaml

@@ -51,6 +51,8 @@ parameter_rules:
     help:
       zh_Hans: 对于每个后续标记，仅从前 K 个选项中进行采样。使用 top_k 删除长尾低概率响应。
       en_US: Only sample from the top K options for each subsequent token. Use top_k to remove long tail low probability responses.
+  - name: response_format
+    use_template: response_format
 pricing:
   input: '0.003'
   output: '0.015'

api/core/model_runtime/model_providers/bedrock/llm/us.anthropic.claude-3-haiku-v1.yaml

@@ -52,6 +52,8 @@ parameter_rules:
     help:
       zh_Hans: 对于每个后续标记，仅从前 K 个选项中进行采样。使用 top_k 删除长尾低概率响应。
       en_US: Only sample from the top K options for each subsequent token. Use top_k to remove long tail low probability responses.
+  - name: response_format
+    use_template: response_format
 pricing:
   input: '0.00025'
   output: '0.00125'

api/core/model_runtime/model_providers/bedrock/llm/us.anthropic.claude-3-opus-v1.yaml

@@ -52,6 +52,8 @@ parameter_rules:
     help:
       zh_Hans: 对于每个后续标记，仅从前 K 个选项中进行采样。使用 top_k 删除长尾低概率响应。
       en_US: Only sample from the top K options for each subsequent token. Use top_k to remove long tail low probability responses.
+  - name: response_format
+    use_template: response_format
 pricing:
   input: '0.015'
   output: '0.075'

api/core/model_runtime/model_providers/bedrock/llm/us.anthropic.claude-3-sonnet-v1.5.yaml

@@ -51,6 +51,8 @@ parameter_rules:
     help:
       zh_Hans: 对于每个后续标记，仅从前 K 个选项中进行采样。使用 top_k 删除长尾低概率响应。
       en_US: Only sample from the top K options for each subsequent token. Use top_k to remove long tail low probability responses.
+  - name: response_format
+    use_template: response_format
 pricing:
   input: '0.003'
   output: '0.015'

api/core/model_runtime/model_providers/bedrock/llm/us.anthropic.claude-3-sonnet-v1.yaml

@@ -51,6 +51,8 @@ parameter_rules:
     help:
       zh_Hans: 对于每个后续标记，仅从前 K 个选项中进行采样。使用 top_k 删除长尾低概率响应。
       en_US: Only sample from the top K options for each subsequent token. Use top_k to remove long tail low probability responses.
+  - name: response_format
+    use_template: response_format
 pricing:
   input: '0.003'
   output: '0.015'

api/core/model_runtime/model_providers/bedrock/text_embedding/text_embedding.py

@@ -13,7 +13,7 @@ from botocore.exceptions import (
     UnknownServiceError,
 )
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.model_entities import PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult
 from core.model_runtime.errors.invoke import (

api/core/model_runtime/model_providers/cohere/text_embedding/text_embedding.py

@@ -5,7 +5,7 @@ import cohere
 import numpy as np
 from cohere.core import RequestOptions
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.model_entities import PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult
 from core.model_runtime.errors.invoke import (

api/core/model_runtime/model_providers/fireworks/fireworks.yaml

@@ -18,6 +18,7 @@ supported_model_types:
   - text-embedding
 configurate_methods:
   - predefined-model
+  - customizable-model
 provider_credential_schema:
   credential_form_schemas:
     - variable: fireworks_api_key
@@ -28,3 +29,75 @@ provider_credential_schema:
       placeholder:
         zh_Hans: 在此输入您的 API Key
         en_US: Enter your API Key
+model_credential_schema:
+  model:
+    label:
+      en_US: Model URL
+      zh_Hans: 模型URL
+    placeholder:
+      en_US: Enter your Model URL
+      zh_Hans: 输入模型URL
+  credential_form_schemas:
+    - variable: model_label_zh_Hanns
+      label:
+        zh_Hans: 模型中文名称
+        en_US: The zh_Hans of Model
+      required: true
+      type: text-input
+      placeholder:
+        zh_Hans: 在此输入您的模型中文名称
+        en_US: Enter your zh_Hans of Model
+    - variable: model_label_en_US
+      label:
+        zh_Hans: 模型英文名称
+        en_US: The en_US of Model
+      required: true
+      type: text-input
+      placeholder:
+        zh_Hans: 在此输入您的模型英文名称
+        en_US: Enter your en_US of Model
+    - variable: fireworks_api_key
+      label:
+        en_US: API Key
+      type: secret-input
+      required: true
+      placeholder:
+        zh_Hans: 在此输入您的 API Key
+        en_US: Enter your API Key
+    - variable: context_size
+      label:
+        zh_Hans: 模型上下文长度
+        en_US: Model context size
+      required: true
+      type: text-input
+      default: '4096'
+      placeholder:
+        zh_Hans: 在此输入您的模型上下文长度
+        en_US: Enter your Model context size
+    - variable: max_tokens
+      label:
+        zh_Hans: 最大 token 上限
+        en_US: Upper bound for max tokens
+      default: '4096'
+      type: text-input
+      show_on:
+        - variable: __model_type
+          value: llm
+    - variable: function_calling_type
+      label:
+        en_US: Function calling
+      type: select
+      required: false
+      default: no_call
+      options:
+        - value: no_call
+          label:
+            en_US: Not Support
+            zh_Hans: 不支持
+        - value: function_call
+          label:
+            en_US: Support
+            zh_Hans: 支持
+      show_on:
+        - variable: __model_type
+          value: llm

api/core/model_runtime/model_providers/fireworks/llm/gemma2-9b-it.yaml

@@ -43,3 +43,4 @@ pricing:
   output: '0.2'
   unit: '0.000001'
   currency: USD
+deprecated: true

api/core/model_runtime/model_providers/fireworks/llm/llm.py

@@ -8,7 +8,8 @@ from openai.types.chat.chat_completion_chunk import ChoiceDeltaFunctionCall, Cho
 from openai.types.chat.chat_completion_message import FunctionCall
 
 from core.model_runtime.callbacks.base_callback import Callback
-from core.model_runtime.entities.llm_entities import LLMResult, LLMResultChunk, LLMResultChunkDelta
+from core.model_runtime.entities.common_entities import I18nObject
+from core.model_runtime.entities.llm_entities import LLMMode, LLMResult, LLMResultChunk, LLMResultChunkDelta
 from core.model_runtime.entities.message_entities import (
     AssistantPromptMessage,
     ImagePromptMessageContent,
@@ -20,6 +21,15 @@ from core.model_runtime.entities.message_entities import (
     ToolPromptMessage,
     UserPromptMessage,
 )
+from core.model_runtime.entities.model_entities import (
+    AIModelEntity,
+    FetchFrom,
+    ModelFeature,
+    ModelPropertyKey,
+    ModelType,
+    ParameterRule,
+    ParameterType,
+)
 from core.model_runtime.errors.validate import CredentialsValidateFailedError
 from core.model_runtime.model_providers.__base.large_language_model import LargeLanguageModel
 from core.model_runtime.model_providers.fireworks._common import _CommonFireworks
@@ -608,3 +618,50 @@ class FireworksLargeLanguageModel(_CommonFireworks, LargeLanguageModel):
                     num_tokens += self._get_num_tokens_by_gpt2(required_field)
 
         return num_tokens
+
+    def get_customizable_model_schema(self, model: str, credentials: dict) -> AIModelEntity:
+        return AIModelEntity(
+            model=model,
+            label=I18nObject(
+                en_US=credentials.get("model_label_en_US", model),
+                zh_Hans=credentials.get("model_label_zh_Hanns", model),
+            ),
+            model_type=ModelType.LLM,
+            features=[ModelFeature.TOOL_CALL, ModelFeature.MULTI_TOOL_CALL, ModelFeature.STREAM_TOOL_CALL]
+            if credentials.get("function_calling_type") == "function_call"
+            else [],
+            fetch_from=FetchFrom.CUSTOMIZABLE_MODEL,
+            model_properties={
+                ModelPropertyKey.CONTEXT_SIZE: int(credentials.get("context_size", 4096)),
+                ModelPropertyKey.MODE: LLMMode.CHAT.value,
+            },
+            parameter_rules=[
+                ParameterRule(
+                    name="temperature",
+                    use_template="temperature",
+                    label=I18nObject(en_US="Temperature", zh_Hans="温度"),
+                    type=ParameterType.FLOAT,
+                ),
+                ParameterRule(
+                    name="max_tokens",
+                    use_template="max_tokens",
+                    default=512,
+                    min=1,
+                    max=int(credentials.get("max_tokens", 4096)),
+                    label=I18nObject(en_US="Max Tokens", zh_Hans="最大标记"),
+                    type=ParameterType.INT,
+                ),
+                ParameterRule(
+                    name="top_p",
+                    use_template="top_p",
+                    label=I18nObject(en_US="Top P", zh_Hans="Top P"),
+                    type=ParameterType.FLOAT,
+                ),
+                ParameterRule(
+                    name="top_k",
+                    use_template="top_k",
+                    label=I18nObject(en_US="Top K", zh_Hans="Top K"),
+                    type=ParameterType.FLOAT,
+                ),
+            ],
+        )

api/core/model_runtime/model_providers/fireworks/llm/qwen2p5-72b-instruct.yaml

@@ -0,0 +1,46 @@
+model: accounts/fireworks/models/qwen2p5-72b-instruct
+label:
+  zh_Hans: Qwen2.5 72B Instruct
+  en_US: Qwen2.5 72B Instruct
+model_type: llm
+features:
+  - agent-thought
+  - tool-call
+model_properties:
+  mode: chat
+  context_size: 32768
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+  - name: top_p
+    use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+  - name: max_tokens
+    use_template: max_tokens
+  - name: context_length_exceeded_behavior
+    default: None
+    label:
+      zh_Hans: 上下文长度超出行为
+      en_US: Context Length Exceeded Behavior
+    help:
+      zh_Hans: 上下文长度超出行为
+      en_US: Context Length Exceeded Behavior
+    type: string
+    options:
+      - None
+      - truncate
+      - error
+  - name: response_format
+    use_template: response_format
+pricing:
+  input: '0.9'
+  output: '0.9'
+  unit: '0.000001'
+  currency: USD

api/core/model_runtime/model_providers/fireworks/text_embedding/text_embedding.py

@@ -5,7 +5,7 @@ from typing import Optional, Union
 import numpy as np
 from openai import OpenAI
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.model_entities import PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult
 from core.model_runtime.errors.validate import CredentialsValidateFailedError

api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-001.yaml

@@ -32,15 +32,6 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-002.yaml

@@ -32,15 +32,6 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-8b-exp-0827.yaml

@@ -32,15 +32,6 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-8b-exp-0924.yaml

@@ -32,15 +32,6 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-exp-0827.yaml

@@ -32,15 +32,6 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash-latest.yaml

@@ -32,15 +32,6 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-1.5-flash.yaml

@@ -32,15 +32,6 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-001.yaml

@@ -32,15 +32,6 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-002.yaml

@@ -32,15 +32,6 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-exp-0801.yaml

@@ -32,15 +32,6 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-exp-0827.yaml

@@ -32,15 +32,6 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro-latest.yaml

@@ -32,15 +32,6 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-1.5-pro.yaml

@@ -32,15 +32,6 @@ parameter_rules:
     max: 8192
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-pro-vision.yaml

@@ -27,15 +27,6 @@ parameter_rules:
     default: 4096
     min: 1
     max: 4096
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/google/llm/gemini-pro.yaml

@@ -31,15 +31,6 @@ parameter_rules:
     max: 2048
   - name: response_format
     use_template: response_format
-  - name: stream
-    label:
-      zh_Hans: 流式输出
-      en_US: Stream
-    type: boolean
-    help:
-      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
-      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
-    default: false
 pricing:
   input: '0.00'
   output: '0.00'

api/core/model_runtime/model_providers/groq/groq.yaml

@@ -18,6 +18,7 @@ help:
     en_US: https://console.groq.com/
 supported_model_types:
   - llm
+  - speech2text
 configurate_methods:
   - predefined-model
 provider_credential_schema:

api/core/model_runtime/model_providers/groq/llm/llama-3.2-11b-vision-preview.yaml

@@ -0,0 +1,26 @@
+model: llama-3.2-11b-vision-preview
+label:
+  zh_Hans: Llama 3.2 11B Vision (Preview)
+  en_US: Llama 3.2 11B Vision (Preview)
+model_type: llm
+features:
+  - agent-thought
+  - vision
+model_properties:
+  mode: chat
+  context_size: 131072
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+  - name: top_p
+    use_template: top_p
+  - name: max_tokens
+    use_template: max_tokens
+    default: 512
+    min: 1
+    max: 8192
+pricing:
+  input: '0.05'
+  output: '0.1'
+  unit: '0.000001'
+  currency: USD

api/core/model_runtime/model_providers/groq/llm/llama-3.2-90b-vision-preview.yaml

@@ -0,0 +1,26 @@
+model: llama-3.2-90b-vision-preview
+label:
+  zh_Hans: Llama 3.2 90B Vision (Preview)
+  en_US: Llama 3.2 90B Vision (Preview)
+model_type: llm
+features:
+  - agent-thought
+  - vision
+model_properties:
+  mode: chat
+  context_size: 131072
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+  - name: top_p
+    use_template: top_p
+  - name: max_tokens
+    use_template: max_tokens
+    default: 512
+    min: 1
+    max: 8192
+pricing:
+  input: '0.05'
+  output: '0.1'
+  unit: '0.000001'
+  currency: USD

api/core/model_runtime/model_providers/groq/speech2text/distil-whisper-large-v3-en.yaml

@@ -0,0 +1,5 @@
+model: distil-whisper-large-v3-en
+model_type: speech2text
+model_properties:
+  file_upload_limit: 1
+  supported_file_extensions: flac,mp3,mp4,mpeg,mpga,m4a,ogg,wav,webm

api/core/model_runtime/model_providers/groq/speech2text/speech2text.py

@@ -0,0 +1,30 @@
+from typing import IO, Optional
+
+from core.model_runtime.model_providers.openai_api_compatible.speech2text.speech2text import OAICompatSpeech2TextModel
+
+
+class GroqSpeech2TextModel(OAICompatSpeech2TextModel):
+    """
+    Model class for Groq Speech to text model.
+    """
+
+    def _invoke(self, model: str, credentials: dict, file: IO[bytes], user: Optional[str] = None) -> str:
+        """
+        Invoke speech2text model
+
+        :param model: model name
+        :param credentials: model credentials
+        :param file: audio file
+        :param user: unique user id
+        :return: text for given audio file
+        """
+        self._add_custom_parameters(credentials)
+        return super()._invoke(model, credentials, file)
+
+    def validate_credentials(self, model: str, credentials: dict) -> None:
+        self._add_custom_parameters(credentials)
+        return super().validate_credentials(model, credentials)
+
+    @classmethod
+    def _add_custom_parameters(cls, credentials: dict) -> None:
+        credentials["endpoint_url"] = "https://api.groq.com/openai/v1"

api/core/model_runtime/model_providers/groq/speech2text/whisper-large-v3-turbo.yaml

@@ -0,0 +1,5 @@
+model: whisper-large-v3-turbo
+model_type: speech2text
+model_properties:
+  file_upload_limit: 1
+  supported_file_extensions: flac,mp3,mp4,mpeg,mpga,m4a,ogg,wav,webm

api/core/model_runtime/model_providers/groq/speech2text/whisper-large-v3.yaml

@@ -0,0 +1,5 @@
+model: whisper-large-v3
+model_type: speech2text
+model_properties:
+  file_upload_limit: 1
+  supported_file_extensions: flac,mp3,mp4,mpeg,mpga,m4a,ogg,wav,webm

api/core/model_runtime/model_providers/huggingface_hub/text_embedding/text_embedding.py

@@ -6,7 +6,7 @@ import numpy as np
 import requests
 from huggingface_hub import HfApi, InferenceClient
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.common_entities import I18nObject
 from core.model_runtime.entities.model_entities import AIModelEntity, FetchFrom, ModelType, PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult

api/core/model_runtime/model_providers/huggingface_tei/text_embedding/text_embedding.py

@@ -1,7 +1,7 @@
 import time
 from typing import Optional
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.common_entities import I18nObject
 from core.model_runtime.entities.model_entities import AIModelEntity, FetchFrom, ModelPropertyKey, ModelType, PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult

api/core/model_runtime/model_providers/hunyuan/text_embedding/text_embedding.py

@@ -9,7 +9,7 @@ from tencentcloud.common.profile.client_profile import ClientProfile
 from tencentcloud.common.profile.http_profile import HttpProfile
 from tencentcloud.hunyuan.v20230901 import hunyuan_client, models
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.model_entities import PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult
 from core.model_runtime.errors.invoke import (

api/core/model_runtime/model_providers/jina/text_embedding/text_embedding.py

@@ -4,7 +4,7 @@ from typing import Optional
 
 from requests import post
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.common_entities import I18nObject
 from core.model_runtime.entities.model_entities import AIModelEntity, FetchFrom, ModelPropertyKey, ModelType, PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult

api/core/model_runtime/model_providers/localai/text_embedding/text_embedding.py

@@ -5,7 +5,7 @@ from typing import Optional
 from requests import post
 from yarl import URL
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.common_entities import I18nObject
 from core.model_runtime.entities.model_entities import AIModelEntity, FetchFrom, ModelPropertyKey, ModelType, PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult

api/core/model_runtime/model_providers/minimax/llm/abab6.5t-chat.yaml

@@ -0,0 +1,44 @@
+model: abab6.5t-chat
+label:
+  en_US: Abab6.5t-Chat
+model_type: llm
+features:
+  - agent-thought
+model_properties:
+  mode: chat
+  context_size: 8192
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+    min: 0.01
+    max: 1
+    default: 0.9
+  - name: top_p
+    use_template: top_p
+    min: 0.01
+    max: 1
+    default: 0.95
+  - name: max_tokens
+    use_template: max_tokens
+    required: true
+    default: 3072
+    min: 1
+    max: 8192
+  - name: mask_sensitive_info
+    type: boolean
+    default: true
+    label:
+      zh_Hans: 隐私保护
+      en_US: Moderate
+    help:
+      zh_Hans: 对输出中易涉及隐私问题的文本信息进行打码，目前包括但不限于邮箱、域名、链接、证件号、家庭住址等，默认true，即开启打码
+      en_US: Mask the sensitive info of the generated content, such as email/domain/link/address/phone/id..
+  - name: presence_penalty
+    use_template: presence_penalty
+  - name: frequency_penalty
+    use_template: frequency_penalty
+pricing:
+  input: '0.005'
+  output: '0.005'
+  unit: '0.001'
+  currency: RMB

api/core/model_runtime/model_providers/minimax/text_embedding/text_embedding.py

@@ -4,7 +4,7 @@ from typing import Optional
 
 from requests import post
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.model_entities import PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult
 from core.model_runtime.errors.invoke import (
@@ -61,7 +61,8 @@ class MinimaxTextEmbeddingModel(TextEmbeddingModel):
         url = f"{self.api_base}?GroupId={group_id}"
         headers = {"Authorization": "Bearer " + api_key, "Content-Type": "application/json"}
 
-        data = {"model": "embo-01", "texts": texts, "type": "db"}
+        embedding_type = "db" if input_type == EmbeddingInputType.DOCUMENT else "query"
+        data = {"model": "embo-01", "texts": texts, "type": embedding_type}
 
         try:
             response = post(url, headers=headers, data=dumps(data))

api/core/model_runtime/model_providers/mixedbread/text_embedding/text_embedding.py

@@ -4,7 +4,7 @@ from typing import Optional
 
 import requests
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.common_entities import I18nObject
 from core.model_runtime.entities.model_entities import AIModelEntity, FetchFrom, ModelPropertyKey, ModelType, PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult

api/core/model_runtime/model_providers/nomic/text_embedding/text_embedding.py

@@ -5,7 +5,7 @@ from typing import Optional
 from nomic import embed
 from nomic import login as nomic_login
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.model_entities import PriceType
 from core.model_runtime.entities.text_embedding_entities import (
     EmbeddingUsage,

api/core/model_runtime/model_providers/nvidia/text_embedding/text_embedding.py

@@ -4,7 +4,7 @@ from typing import Optional
 
 from requests import post
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.model_entities import PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult
 from core.model_runtime.errors.invoke import (

api/core/model_runtime/model_providers/oci/text_embedding/text_embedding.py

@@ -6,7 +6,7 @@ from typing import Optional
 import numpy as np
 import oci
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.model_entities import PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult
 from core.model_runtime.errors.invoke import (

api/core/model_runtime/model_providers/ollama/text_embedding/text_embedding.py

@@ -8,7 +8,7 @@ from urllib.parse import urljoin
 import numpy as np
 import requests
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.common_entities import I18nObject
 from core.model_runtime.entities.model_entities import (
     AIModelEntity,

api/core/model_runtime/model_providers/openai/openai.py

@@ -19,9 +19,9 @@ class OpenAIProvider(ModelProvider):
         try:
             model_instance = self.get_model_instance(ModelType.LLM)
 
-            # Use `gpt-3.5-turbo` model for validate,
+            # Use `gpt-4o-mini` model for validate,
             # no matter what model you pass in, text completion model or chat model
-            model_instance.validate_credentials(model="gpt-3.5-turbo", credentials=credentials)
+            model_instance.validate_credentials(model="gpt-4o-mini", credentials=credentials)
         except CredentialsValidateFailedError as ex:
             raise ex
         except Exception as ex:

api/core/model_runtime/model_providers/openai/text_embedding/text_embedding.py

@@ -6,7 +6,7 @@ import numpy as np
 import tiktoken
 from openai import OpenAI
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.model_entities import PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult
 from core.model_runtime.errors.validate import CredentialsValidateFailedError

api/core/model_runtime/model_providers/openai_api_compatible/openai_api_compatible.yaml

@@ -8,6 +8,7 @@ supported_model_types:
   - llm
   - text-embedding
   - speech2text
+  - rerank
 configurate_methods:
   - customizable-model
 model_credential_schema:
@@ -83,6 +84,19 @@ model_credential_schema:
       placeholder:
         zh_Hans: 在此输入您的模型上下文长度
         en_US: Enter your Model context size
+    - variable: context_size
+      label:
+        zh_Hans: 模型上下文长度
+        en_US: Model context size
+      required: true
+      show_on:
+        - variable: __model_type
+          value: rerank
+      type: text-input
+      default: '4096'
+      placeholder:
+        zh_Hans: 在此输入您的模型上下文长度
+        en_US: Enter your Model context size
     - variable: max_tokens_to_sample
       label:
         zh_Hans: 最大 token 上限

api/core/model_runtime/model_providers/openai_api_compatible/rerank/rerank.py

@@ -0,0 +1,159 @@
+from json import dumps
+from typing import Optional
+
+import httpx
+from requests import post
+from yarl import URL
+
+from core.model_runtime.entities.common_entities import I18nObject
+from core.model_runtime.entities.model_entities import AIModelEntity, FetchFrom, ModelType
+from core.model_runtime.entities.rerank_entities import RerankDocument, RerankResult
+from core.model_runtime.errors.invoke import (
+    InvokeAuthorizationError,
+    InvokeBadRequestError,
+    InvokeConnectionError,
+    InvokeError,
+    InvokeRateLimitError,
+    InvokeServerUnavailableError,
+)
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+from core.model_runtime.model_providers.__base.rerank_model import RerankModel
+
+
+class OAICompatRerankModel(RerankModel):
+    """
+    rerank model API is compatible with Jina rerank model API. So copy the JinaRerankModel class code here.
+    we need enhance for llama.cpp , which return raw score, not normalize score 0~1.  It seems Dify need it
+    """
+
+    def _invoke(
+        self,
+        model: str,
+        credentials: dict,
+        query: str,
+        docs: list[str],
+        score_threshold: Optional[float] = None,
+        top_n: Optional[int] = None,
+        user: Optional[str] = None,
+    ) -> RerankResult:
+        """
+        Invoke rerank model
+
+        :param model: model name
+        :param credentials: model credentials
+        :param query: search query
+        :param docs: docs for reranking
+        :param score_threshold: score threshold
+        :param top_n: top n documents to return
+        :param user: unique user id
+        :return: rerank result
+        """
+        if len(docs) == 0:
+            return RerankResult(model=model, docs=[])
+
+        server_url = credentials["endpoint_url"]
+        model_name = model
+
+        if not server_url:
+            raise CredentialsValidateFailedError("server_url is required")
+        if not model_name:
+            raise CredentialsValidateFailedError("model_name is required")
+
+        url = server_url
+        headers = {"Authorization": f"Bearer {credentials.get('api_key')}", "Content-Type": "application/json"}
+
+        # TODO: Do we need truncate docs to avoid llama.cpp return error?
+
+        data = {"model": model_name, "query": query, "documents": docs, "top_n": top_n}
+
+        try:
+            response = post(str(URL(url) / "rerank"), headers=headers, data=dumps(data), timeout=60)
+            response.raise_for_status()
+            results = response.json()
+
+            rerank_documents = []
+            scores = [result["relevance_score"] for result in results["results"]]
+
+            # Min-Max Normalization: Normalize scores to 0 ~ 1.0 range
+            min_score = min(scores)
+            max_score = max(scores)
+            score_range = max_score - min_score if max_score != min_score else 1.0  # Avoid division by zero
+
+            for result in results["results"]:
+                index = result["index"]
+
+                # Retrieve document text (fallback if llama.cpp rerank doesn't return it)
+                text = result.get("document", {}).get("text", docs[index])
+
+                # Normalize the score
+                normalized_score = (result["relevance_score"] - min_score) / score_range
+
+                # Create RerankDocument object with normalized score
+                rerank_document = RerankDocument(
+                    index=index,
+                    text=text,
+                    score=normalized_score,
+                )
+
+                # Apply threshold (if defined)
+                if score_threshold is None or normalized_score >= score_threshold:
+                    rerank_documents.append(rerank_document)
+
+            # Sort rerank_documents by normalized score in descending order
+            rerank_documents.sort(key=lambda doc: doc.score, reverse=True)
+
+            return RerankResult(model=model, docs=rerank_documents)
+
+        except httpx.HTTPStatusError as e:
+            raise InvokeServerUnavailableError(str(e))
+
+    def validate_credentials(self, model: str, credentials: dict) -> None:
+        """
+        Validate model credentials
+
+        :param model: model name
+        :param credentials: model credentials
+        :return:
+        """
+        try:
+            self._invoke(
+                model=model,
+                credentials=credentials,
+                query="What is the capital of the United States?",
+                docs=[
+                    "Carson City is the capital city of the American state of Nevada. At the 2010 United States "
+                    "Census, Carson City had a population of 55,274.",
+                    "The Commonwealth of the Northern Mariana Islands is a group of islands in the Pacific Ocean that "
+                    "are a political division controlled by the United States. Its capital is Saipan.",
+                ],
+                score_threshold=0.8,
+            )
+        except Exception as ex:
+            raise CredentialsValidateFailedError(str(ex))
+
+    @property
+    def _invoke_error_mapping(self) -> dict[type[InvokeError], list[type[Exception]]]:
+        """
+        Map model invoke error to unified error
+        """
+        return {
+            InvokeConnectionError: [httpx.ConnectError],
+            InvokeServerUnavailableError: [httpx.RemoteProtocolError],
+            InvokeRateLimitError: [],
+            InvokeAuthorizationError: [httpx.HTTPStatusError],
+            InvokeBadRequestError: [httpx.RequestError],
+        }
+
+    def get_customizable_model_schema(self, model: str, credentials: dict) -> AIModelEntity:
+        """
+        generate custom model entities from credentials
+        """
+        entity = AIModelEntity(
+            model=model,
+            label=I18nObject(en_US=model),
+            model_type=ModelType.RERANK,
+            fetch_from=FetchFrom.CUSTOMIZABLE_MODEL,
+            model_properties={},
+        )
+
+        return entity

api/core/model_runtime/model_providers/openai_api_compatible/text_embedding/text_embedding.py

@@ -7,7 +7,7 @@ from urllib.parse import urljoin
 import numpy as np
 import requests
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.common_entities import I18nObject
 from core.model_runtime.entities.model_entities import (
     AIModelEntity,

api/core/model_runtime/model_providers/openllm/text_embedding/text_embedding.py

@@ -5,7 +5,7 @@ from typing import Optional
 from requests import post
 from requests.exceptions import ConnectionError, InvalidSchema, MissingSchema
 
-from core.embedding.embedding_constant import EmbeddingInputType
+from core.entities.embedding_type import EmbeddingInputType
 from core.model_runtime.entities.model_entities import PriceType
 from core.model_runtime.entities.text_embedding_entities import EmbeddingUsage, TextEmbeddingResult
 from core.model_runtime.errors.invoke import (

api/core/model_runtime/model_providers/openrouter/llm/deepseek-chat.yaml

@@ -35,6 +35,15 @@ parameter_rules:
     help:
       zh_Hans: 控制生成结果的随机性。数值越小，随机性越弱；数值越大，随机性越强。一般而言，top_p 和 temperature 两个参数选择一个进行调整即可。
       en_US: Control the randomness of generated results. The smaller the value, the weaker the randomness; the larger the value, the stronger the randomness. Generally speaking, you can adjust one of the two parameters top_p and temperature.
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+    required: false
   - name: frequency_penalty
     use_template: frequency_penalty
     default: 0

api/core/model_runtime/model_providers/openrouter/llm/deepseek-coder.yaml

@@ -18,6 +18,15 @@ parameter_rules:
     min: 0
     max: 1
     default: 1
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+    required: false
   - name: max_tokens
     use_template: max_tokens
     min: 1

api/core/model_runtime/model_providers/openrouter/llm/gpt-3.5-turbo.yaml

@@ -14,6 +14,15 @@ parameter_rules:
     use_template: temperature
   - name: top_p
     use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+    required: false
   - name: presence_penalty
     use_template: presence_penalty
   - name: frequency_penalty

api/core/model_runtime/model_providers/openrouter/llm/gpt-4-32k.yaml

@@ -14,6 +14,15 @@ parameter_rules:
     use_template: temperature
   - name: top_p
     use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+    required: false
   - name: presence_penalty
     use_template: presence_penalty
   - name: frequency_penalty

api/core/model_runtime/model_providers/openrouter/llm/gpt-4.yaml

@@ -14,6 +14,15 @@ parameter_rules:
     use_template: temperature
   - name: top_p
     use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+    required: false
   - name: presence_penalty
     use_template: presence_penalty
   - name: frequency_penalty

api/core/model_runtime/model_providers/openrouter/llm/gpt-4o-2024-08-06.yaml

@@ -16,6 +16,15 @@ parameter_rules:
     use_template: temperature
   - name: top_p
     use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+    required: false
   - name: presence_penalty
     use_template: presence_penalty
   - name: frequency_penalty

api/core/model_runtime/model_providers/openrouter/llm/gpt-4o-mini.yaml

@@ -15,6 +15,15 @@ parameter_rules:
     use_template: temperature
   - name: top_p
     use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+    required: false
   - name: presence_penalty
     use_template: presence_penalty
   - name: frequency_penalty

api/core/model_runtime/model_providers/openrouter/llm/gpt-4o.yaml

@@ -15,6 +15,15 @@ parameter_rules:
     use_template: temperature
   - name: top_p
     use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+    required: false
   - name: presence_penalty
     use_template: presence_penalty
   - name: frequency_penalty

api/core/model_runtime/model_providers/openrouter/llm/llama-3-70b-instruct.yaml

@@ -10,6 +10,15 @@ parameter_rules:
     use_template: temperature
   - name: top_p
     use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+    required: false
   - name: max_tokens
     use_template: max_tokens
     required: true