improve api docs

backend-python/routes/completion.py

@@ -24,6 +24,21 @@ class ChatCompletionBody(ModelConfigBody):
     stream: bool = False
     stop: str = None
 
+    class Config:
+        schema_extra = {
+            "example": {
+                "messages": [{"role": "user", "content": "hello"}],
+                "model": "rwkv",
+                "stream": False,
+                "stop": None,
+                "max_tokens": 1000,
+                "temperature": 1.2,
+                "top_p": 0.5,
+                "presence_penalty": 0.4,
+                "frequency_penalty": 0.4,
+            }
+        }
+
 
 completion_lock = Lock()
 
@@ -242,6 +257,21 @@ class CompletionBody(ModelConfigBody):
     stream: bool = False
     stop: str = None
 
+    class Config:
+        schema_extra = {
+            "example": {
+                "prompt": "The following is an epic science fiction masterpiece that is immortalized, with delicate descriptions and grand depictions of interstellar civilization wars.\nChapter 1.\n",
+                "model": "rwkv",
+                "stream": False,
+                "stop": None,
+                "max_tokens": 100,
+                "temperature": 1.2,
+                "top_p": 0.5,
+                "presence_penalty": 0.4,
+                "frequency_penalty": 0.4,
+            }
+        }
+
 
 @router.post("/v1/completions")
 @router.post("/completions")

backend-python/routes/config.py

@@ -29,6 +29,15 @@ class SwitchModelBody(BaseModel):
     strategy: str
     customCuda: bool = False
 
+    class Config:
+        schema_extra = {
+            "example": {
+                "model": "models/RWKV-4-World-3B-v1-OnlyForTest_80%_trained-20230612-ctx4096.pth",
+                "strategy": "cuda fp16",
+                "customCuda": False,
+            }
+        }
+
 
 @router.post("/switch-model")
 def switch_model(body: SwitchModelBody, response: Response, request: Request):
@@ -59,7 +68,9 @@ def switch_model(body: SwitchModelBody, response: Response, request: Request):
         print(e)
         quick_log(request, body, f"Exception: {e}")
         global_var.set(global_var.Model_Status, global_var.ModelStatus.Offline)
-        raise HTTPException(Status.HTTP_500_INTERNAL_SERVER_ERROR, "failed to load")
+        raise HTTPException(
+            Status.HTTP_500_INTERNAL_SERVER_ERROR, f"failed to load: {e}"
+        )
 
     if global_var.get(global_var.Model_Config) is None:
         global_var.set(