🐛 bugfix: model status sync, SiliconFlow logic alignment, and VLM connectivity

backend/services/model_management_service.py

@@ -199,7 +199,11 @@ async def list_provider_models_for_tenant(tenant_id: str, provider: str, model_t
         model_list = get_models_by_tenant_factory_type(
             tenant_id, provider, model_type)
         for model in model_list:
-            model["id"] = model["model_repo"] + "/" + model["model_name"]
+            # Use add_repo_to_name for consistent format with /model/list API
+            model["id"] = add_repo_to_name(
+                model_repo=model["model_repo"],
+                model_name=model["model_name"],
+            )
 
         logging.debug(f"Provider model {provider} created successfully")
         return model_list

frontend/app/[locale]/models/components/model/ModelAddDialog.tsx

@@ -1300,19 +1300,47 @@ export const ModelAddDialog = ({
                 </a>
               </Tooltip>
               {form.isBatchImport && (
-                <Tooltip title="SiliconFlow">
-                  <a
-                    href={PROVIDER_LINKS.siliconflow}
-                    target="_blank"
-                    rel="noopener noreferrer"
-                  >
-                    <img
-                      src="/siliconflow.png"
-                      alt="SiliconFlow"
-                      className="h-4 ml-1.5 cursor-pointer"
-                    />
-                  </a>
-                </Tooltip>
+                <>
+                  <Tooltip title="SiliconFlow">
+                    <a
+                      href={PROVIDER_LINKS.siliconflow}
+                      target="_blank"
+                      rel="noopener noreferrer"
+                    >
+                      <img
+                        src="/siliconflow.png"
+                        alt="SiliconFlow"
+                        className="h-4 ml-1.5 cursor-pointer"
+                      />
+                    </a>
+                  </Tooltip>
+                  <Tooltip title={t("model.provider.dashscope")}>
+                    <a
+                      href={PROVIDER_LINKS.dashscope}
+                      target="_blank"
+                      rel="noopener noreferrer"
+                    >
+                      <img
+                        src="/aliyuncs.png"
+                        alt="DashScope"
+                        className="h-4 ml-1.5 cursor-pointer"
+                      />
+                    </a>
+                  </Tooltip>
+                  <Tooltip title={t("model.provider.tokenpony")}>
+                    <a
+                      href={PROVIDER_LINKS.tokenpony}
+                      target="_blank"
+                      rel="noopener noreferrer"
+                    >
+                      <img
+                        src="/tokenpony.png"
+                        alt="TokenPony"
+                        className="h-4 ml-1.5 cursor-pointer"
+                      />
+                    </a>
+                  </Tooltip>
+                </>
               )}
               {form.type === "llm" && !form.isBatchImport && (
                 <>

frontend/app/[locale]/models/components/model/ModelDeleteDialog.tsx

@@ -1252,7 +1252,9 @@ export const ModelDeleteDialog = ({
           </div>
 
           {(selectedSource === MODEL_SOURCES.SILICON ||
-            selectedSource === MODEL_SOURCES.MODELENGINE) &&
+            selectedSource === MODEL_SOURCES.MODELENGINE ||
+            selectedSource === MODEL_SOURCES.DASHSCOPE ||
+            selectedSource === MODEL_SOURCES.TOKENPONY) &&
           providerModels.length > 0 ? (
             <div className="max-h-60 overflow-y-auto border border-gray-200 rounded-md divide-y divide-gray-200">
               {providerModels.length > 0 && (

frontend/const/modelConfig.ts

@@ -84,7 +84,9 @@ export const PROVIDER_LINKS: Record<string, string> = {
   deepseek: "https://platform.deepseek.com/",
   qwen: "https://bailian.console.aliyun.com/",
   jina: "https://jina.ai/",
-  baai: "https://www.baai.ac.cn/"
+  baai: "https://www.baai.ac.cn/",
+  dashscope: "https://dashscope.aliyun.com/",
+  tokenpony: "https://www.tokenpony.cn/"
 };
 
 // User role constants

sdk/nexent/core/models/openai_vlm.py

@@ -1,4 +1,6 @@
+import asyncio
 import base64
+import logging
 import os
 from typing import List, Dict, Any, Union, BinaryIO
 
@@ -7,6 +9,8 @@
 from ..models import OpenAIModel
 from ..utils.observer import MessageObserver
 
+logger = logging.getLogger(__name__)
+
 
 class OpenAIVLModel(OpenAIModel):
     def __init__(
@@ -32,17 +36,48 @@ def __init__(
 
     async def check_connectivity(self) -> bool:
         """
-        Check the connectivity of the VLM model.
+        Check the connectivity of the VLM model by sending a test request with
+        a text prompt and an image. VLM APIs (especially DashScope qwen-vl)
+        require specific format: content as a list with 'type': 'image' and
+        'type': 'text' objects.
 
         Returns:
-            bool: Returns True if the model can respond normally, otherwise returns False.
+            bool: True if the model responds successfully, otherwise False.
         """
+        # Use local test image from images folder - use absolute path based on module location
+        module_dir = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+        test_image_path = os.path.join(module_dir, "assets", "git-flow.png")
+        if os.path.exists(test_image_path):
+            base64_image = self.encode_image(test_image_path)
+            # Detect image format for proper MIME type
+            _, ext = os.path.splitext(test_image_path)
+            image_format = ext.lower()[1:] if ext else "png"
+            if image_format == "jpg":
+                image_format = "jpeg"
+
+            content_parts: List[Dict[str, Any]] = [
+                {"type": "image_url", "image_url": {"url": f"data:image/{image_format};base64,{base64_image}"}},
+                {"type": "text", "text": "Hello"},
+            ]
+        else:
+            # Fallback to remote URL if local image not found
+            test_image_url = "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20250925/thtclx/input1.png"
+            content_parts = [
+                {"type": "image_url", "image_url": {"url": test_image_url}},
+                {"type": "text", "text": "Hello"},
+            ]
+
         try:
-            # Directly reuse the parent class's check_connectivity method
-            return await super().check_connectivity()
+            await asyncio.to_thread(
+                self.client.chat.completions.create,
+                model=self.model_id,
+                messages=[{"role": "user", "content": content_parts}],
+                max_tokens=5,
+                stream=False,
+            )
+            return True
         except Exception as e:
-            import logging
-            logging.error(f"VLM connectivity check failed: {str(e)}")
+            logger.error("VLM connectivity check failed: %s", e)
             return False
 
     def encode_image(self, image_input: Union[str, BinaryIO]) -> str:
@@ -87,7 +122,7 @@ def prepare_image_message(self, image_input: Union[str, BinaryIO], system_prompt
 
         messages = [{"role": "system", "content": [{"text": system_prompt, "type": "text"}]}, {"role": "user",
             "content": [{"type": "image_url",
-                "image_url": {"url": f"data:image/jpeg;base64,{base64_image}", "detail": "auto"}}]}]
+                "image_url": {"url": f"data:image/{image_format};base64,{base64_image}", "detail": "auto"}}]}]
 
         return messages