Update main.py

main.py

@@ -5,10 +5,13 @@ import uvicorn
 from fastapi.responses import HTMLResponse
 from fastapi.middleware.cors import CORSMiddleware
 from sse_starlette.sse import EventSourceResponse
+from ctransformers import AutoModelForCausalLM
 from ctransformers.langchain import CTransformers
 from pydantic import BaseModel
 
-llm = CTransformers(model='ggml-model-q4_1.bin', model_type='starcoder')
+llm = AutoModelForCausalLM.from_pretrained("TheBloke/gorilla-7B-GGML",
+                                           model_file="Gorilla-7B.ggmlv3.q4_0.bin",
+                                           model_type="llama")
 app = fastapi.FastAPI()
 app.add_middleware(
     CORSMiddleware,
@@ -30,13 +33,14 @@ class ChatCompletionRequest(BaseModel):
 
 @app.post("/v1/chat/completions")
 async def chat(request: ChatCompletionRequest, response_mode=None):
-    completion = llm(request.prompt)
-    async def server_sent_events(chat_chunks):
-        for chat_chunk in chat_chunks:
-            yield dict(data=json.dumps(chat_chunk))
-        yield dict(data="[DONE]")
+    tokens = llm.tokenize(prompt)
+    async def server_sent_events(chat_chunks, llm):
+        yield prompt
+        for chat_chunk in llm.generate(chat_chunks):
+            yield llm.detokenize(chat_chunk)
+        yield ""
 
-    return EventSourceResponse(server_sent_events(completion))
+    return EventSourceResponse(server_sent_events(tokens, llm))
 
 if __name__ == "__main__":
   uvicorn.run(app, host="0.0.0.0", port=8000)