Update main.py

main.py

@@ -2,11 +2,12 @@ import fastapi
 import json
 import markdown
 import uvicorn
-from fastapi.responses import StreamingResponse, HTMLResponse
+from fastapi.responses import HTMLResponse
 from fastapi.middleware.cors import CORSMiddleware
 from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
 from ctransformers import AutoModelForCausalLM
 from pydantic import BaseModel
+from sse_starlette.sse import EventSourceResponse
 
 config = {"max_seq_len": 4096}
 llm = AutoModelForCausalLM.from_pretrained('TheBloke/MPT-7B-Storywriter-GGML',
@@ -45,13 +46,20 @@ async def chat(prompt = "Once upon a time there was a "):
 @app.post("/v1/chat/completions")
 async def chat(request: ChatCompletionRequest, response_mode=None):
     completion = llm(request.prompt)
-    async def server_sent_events(chat_chunks):
+
+    async def server_sent_events(
+        chat_chunks,
+    ):
         for chat_chunk in chat_chunks:
-            print(chat_chunk)
-            yield chat_chunk
-        yield "[DONE]"
+            yield dict(data=json.dumps(chat_chunk))
+        yield dict(data="[DONE]")
+
+    chunks = completion_or_chunks
+
+    return EventSourceResponse(
+        server_sent_events(chunks),
+    )
 
-    return StreamingResponse(server_sent_events(completion))
 
 if __name__ == "__main__":
   uvicorn.run(app, host="0.0.0.0", port=8000)