Support downloading models from ModelScope Community. (#5073)

This PR supports downloading models from ModelScope. The main
modifications are as follows:
-New Feature (non-breaking change which adds functionality)
-Documentation Update

---------

Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>

api/apps/llm_app.py

@@ -360,4 +360,4 @@ def list_app():
 
         return get_json_result(data=res)
     except Exception as e:
-        return server_error_response(e)
+        return server_error_response(e)

conf/llm_factories.json

@@ -295,6 +295,13 @@
             "status": "1",
             "llm": []
         },
+        {
+            "name": "ModelScope",
+            "logo": "",
+            "tags": "LLM",
+            "status": "1",
+            "llm": []
+        },
         {
             "name": "LocalAI",
             "logo": "",

docs/references/supported_models.mdx

@@ -35,6 +35,7 @@ A complete list of models supported by RAGFlow, which will continue to expand.
 | LM-Studio             | :heavy_check_mark: | :heavy_check_mark: | :heavy_check_mark: | :heavy_check_mark: |                    |                    |
 | MiniMax               | :heavy_check_mark: |                    |                    |                    |                    |                    |
 | Mistral               | :heavy_check_mark: | :heavy_check_mark: |                    |                    |                    |                    |
+| ModelScope            | :heavy_check_mark: |                    |                    |                    |                    |                    |
 | Moonshot              | :heavy_check_mark: |                    |                    | :heavy_check_mark: |                    |                    |
 | novita.ai             | :heavy_check_mark: |                    |                    |                    |                    |                    |
 | NVIDIA                | :heavy_check_mark: | :heavy_check_mark: | :heavy_check_mark: | :heavy_check_mark: |                    |                    |
@@ -70,4 +71,4 @@ If your model is not listed here but has APIs compatible with those of OpenAI, c
 
 :::note
 The list of supported models is extracted from [this source](https://github.com/infiniflow/ragflow/blob/main/rag/llm/__init__.py) and may not be the most current. For the latest supported model list, please refer to the Python file.
-:::
+:::

rag/llm/__init__.py

@@ -85,6 +85,7 @@ from .chat_model import (
     GoogleChat,
     HuggingFaceChat,
     GPUStackChat,
+    ModelScopeChat,
 )
 
 from .cv_model import (
@@ -231,6 +232,7 @@ ChatModel = {
     "Google Cloud": GoogleChat,
     "HuggingFace": HuggingFaceChat,
     "GPUStack": GPUStackChat,
+    "ModelScope":ModelScopeChat,
 }
 
 RerankModel = {

rag/llm/chat_model.py

@@ -143,6 +143,16 @@ class HuggingFaceChat(Base):
         super().__init__(key, model_name.split("___")[0], base_url)
 
 
+class ModelScopeChat(Base):
+    def __init__(self, key=None, model_name="", base_url=""):
+        if not base_url:
+            raise ValueError("Local llm url cannot be None")
+        base_url = base_url.rstrip('/')
+        if base_url.split("/")[-1] != "v1":
+            base_url = os.path.join(base_url, "v1")
+        super().__init__(key, model_name.split("___")[0], base_url)
+
+
 class DeepSeekChat(Base):
     def __init__(self, key, model_name="deepseek-chat", base_url="https://api.deepseek.com/v1"):
         if not base_url:
@@ -1537,4 +1547,4 @@ class GPUStackChat(Base):
             raise ValueError("Local llm url cannot be None")
         if base_url.split("/")[-1] != "v1-openai":
             base_url = os.path.join(base_url, "v1-openai")
-        super().__init__(key, model_name, base_url)
+        super().__init__(key, model_name, base_url)

web/src/assets/svg/llm/modelscope.svg

@@ -0,0 +1 @@
+<svg id="_层_2" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 274.37 172.76"><defs><style>.cls-2{fill:#36cfd1}.cls-3{fill:#624aff}</style></defs><g id="_层_1-2"><path class="cls-3" d="M24.78 73.55h25.65V99.2H24.78zm99.14 25.66h25.65v25.65h-25.65zm76.95 25.65h-25.65v22.19h47.84V99.21h-22.19v25.65z"/><path class="cls-2" d="M149.57 73.55h25.65V99.2h-25.65zM24.78 47.9h25.65v25.65H24.78z"/><path class="cls-3" d="M223.06 73.55h25.65V99.2h-25.65z"/><path class="cls-2" d="M223.06 47.9h25.65v25.65h-25.65z"/><path class="cls-3" d="M175.22 25.71V47.9h25.65v25.65h22.19V25.71h-47.84z"/><path class="cls-2" d="M98.27 73.55h25.65V99.2H98.27z"/><path class="cls-3" d="M72.62 47.9h25.65V25.71H50.43v47.84h22.19V47.9zm0 51.31H50.43v47.84h47.84v-22.19H72.62V99.21z"/><path style="fill:none" d="M0 0h274.37v172.76H0z"/></g></svg>

web/src/constants/setting.ts

@@ -33,6 +33,7 @@ export const IconMap = {
   文心一言: 'wenxin',
   Ollama: 'ollama',
   Xinference: 'xinference',
+  ModelScope: 'modelscope',
   DeepSeek: 'deepseek',
   VolcEngine: 'volc_engine',
   BaiChuan: 'baichuan',

web/src/pages/user-setting/constants.tsx

@@ -32,6 +32,7 @@ export const LocalLlmFactories = [
   'OpenRouter',
   'HuggingFace',
   'GPUStack',
+  'ModelScope',
 ];
 
 export enum TenantRole {

web/src/pages/user-setting/setting-model/ollama-modal/index.tsx

@@ -21,6 +21,7 @@ const llmFactoryToUrlMap = {
   Ollama:
     'https://github.com/infiniflow/ragflow/blob/main/docs/guides/deploy_local_llm.mdx',
   Xinference: 'https://inference.readthedocs.io/en/latest/user_guide',
+  ModelScope: 'https://www.modelscope.cn/docs/model-service/API-Inference/intro',
   LocalAI: 'https://localai.io/docs/getting-started/models/',
   'LM-Studio': 'https://lmstudio.ai/docs/basics',
   'OpenAI-API-Compatible': 'https://platform.openai.com/docs/models/gpt-4',
@@ -77,6 +78,9 @@ const OllamaModal = ({
       { value: 'speech2text', label: 'sequence2text' },
       { value: 'tts', label: 'tts' },
     ],
+    ModelScope: [
+      { value: 'chat', label: 'chat' },
+    ],
     GPUStack: [
       { value: 'chat', label: 'chat' },
       { value: 'embedding', label: 'embedding' },