Create ollama.py

ollama.py

@@ -0,0 +1,72 @@
+from fastapi import FastAPI, Request, HTTPException
+from fastapi.responses import StreamingResponse
+import httpx
+import logging
+
+app = FastAPI()
+OLLAMA_BASE_URL = "http://localhost:11434"
+client = httpx.AsyncClient(base_url=OLLAMA_BASE_URL)
+
+# Set up logging
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger("uvicorn")
+
+@app.middleware("http")
+async def log_requests(request: Request, call_next):
+    logger.info(f"Request: {request.method} {request.url}")
+    response = await call_next(request)
+    return response
+
+@app.api_route("/{path:path}", methods=["GET", "POST", "PUT", "DELETE"])
+async def reverse_proxy(request: Request, path: str):
+    try:
+        # Build the Ollama URL
+        url = f"{OLLAMA_BASE_URL}/{path}"
+        
+        # Forward headers excluding those that might cause issues
+        headers = {
+            key: value
+            for key, value in request.headers.items()
+            if key.lower() not in ["host", "content-length"]
+        }
+        
+        # Forward the request to Ollama
+        req = client.build_request(
+            method=request.method,
+            url=url,
+            headers=headers,
+            content=await request.body(),
+            params=request.query_params
+        )
+        
+        response = await client.send(req, stream=True)
+        
+        # Handle streaming responses
+        if "text/event-stream" in response.headers.get("content-type", ""):
+            return StreamingResponse(
+                response.aiter_bytes(),
+                media_type=response.headers.get("content-type"),
+                headers=dict(response.headers)
+            )
+            
+        return StreamingResponse(
+            response.aiter_bytes(),
+            media_type=response.headers.get("content-type"),
+            headers=dict(response.headers)
+        )
+        
+    except httpx.ConnectError:
+        raise HTTPException(
+            status_code=503,
+            detail="Ollama server unavailable"
+        )
+    except Exception as e:
+        logger.error(f"Proxy error: {str(e)}")
+        raise HTTPException(
+            status_code=500,
+            detail=f"Proxy error: {str(e)}"
+        )
+
+if __name__ == "__main__":
+    import uvicorn
+    uvicorn.run(app, host="0.0.0.0", port=7860)