Update app.py

app.py

@@ -13,21 +13,28 @@ app = FastAPI()
 client = Client("CohereLabs/command-a-vision")
 
 
-# call gradio safely
-def call_gradio(message, max_tokens=12800, temperature=0.1, top_p=0.9):
+# ✅ FIXED: call gradio with positional args
+def call_gradio(message, max_tokens=100):
 
     try:
-        job = client.submit(
-            message=message,
-            max_tokens=max_tokens,
-            temperature=temperature,
-            top_p=top_p,
+        # format input like Gradio expects
+        payload = {
+            "text": message,
+            "files": []
+        }
+
+        # IMPORTANT: positional inputs (NOT keyword args)
+        result = client.predict(
+            payload,       # input 1
+            max_tokens,    # input 2
             api_name="/chat"
         )
 
-        result = job.result()
+        # result comes as dict sometimes
+        if isinstance(result, dict):
+            return json.dumps(result)
 
-        return result
+        return str(result)
 
     except Exception as e:
         print("Gradio API error:", e)
@@ -39,7 +46,7 @@ def format_openai_response(content):
         "id": f"chatcmpl-{uuid.uuid4().hex}",
         "object": "chat.completion",
         "created": int(time.time()),
-        "model": "minimax-text-01",
+        "model": "command-a-vision",
         "choices": [
             {
                 "index": 0,
@@ -61,23 +68,21 @@ async def chat(request: Request):
     messages = body.get("messages", [])
     stream = body.get("stream", False)
 
-    max_tokens = body.get("max_tokens", 12800)
-    temperature = body.get("temperature", 0.1)
-    top_p = body.get("top_p", 0.9)
+    max_tokens = body.get("max_tokens", 100)
 
     user_message = messages[-1]["content"]
 
-    # normal response
+    # ✅ normal response
     if not stream:
 
-        result = call_gradio(user_message, max_tokens, temperature, top_p)
+        result = call_gradio(user_message, max_tokens)
 
         return JSONResponse(format_openai_response(result))
 
-    # streaming response
+    # ✅ streaming response
     async def generate():
 
-        result = call_gradio(user_message, max_tokens, temperature, top_p)
+        result = call_gradio(user_message, max_tokens)
 
         words = result.split(" ")
 
@@ -87,7 +92,7 @@ async def chat(request: Request):
                 "id": f"chatcmpl-{uuid.uuid4().hex}",
                 "object": "chat.completion.chunk",
                 "created": int(time.time()),
-                "model": "minimax-text-01",
+                "model": "command-a-vision",
                 "choices": [
                     {
                         "delta": {"content": word + " "},
@@ -98,9 +103,9 @@ async def chat(request: Request):
             }
 
             yield f"data: {json.dumps(chunk)}\n\n"
-
             await asyncio.sleep(0.02)
 
+        # end
         end_chunk = {
             "id": f"chatcmpl-{uuid.uuid4().hex}",
             "object": "chat.completion.chunk",