Update app.py

app.py

@@ -44,23 +44,14 @@
 
 
 
-
-
-
-
-
-
-
 import os
-import asyncio
 from fastapi import FastAPI
 from pydantic import BaseModel
-import gradio as gr
 from llama_cpp import Llama
 from huggingface_hub import snapshot_download
 
-# Optionally increase the HTTP timeout via an environment variable
-os.environ["HF_HUB_HTTP_TIMEOUT"] = "3600"  # 1 hour
+# Optionally increase the HTTP timeout via an environment variable (1 hour)
+os.environ["HF_HUB_HTTP_TIMEOUT"] = "3600"
 
 # Download the model repository locally; this call blocks until the download completes.
 local_repo = snapshot_download(
@@ -74,19 +65,20 @@ local_repo = snapshot_download(
 llm = Llama.from_pretrained(
     repo_id="DebabrataHalder/mysqlmodel",  # valid repo id format
     filename="mysqlmodel.gguf",
-    cache_dir=local_repo,                   # point to the downloaded snapshot
-    n_ctx=2048,                             # adjust context size if needed
-    local_files_only=True                   # ensure only local files are used
+    cache_dir=local_repo,  # point to the downloaded snapshot
+    n_ctx=2048,          # adjust context size if needed
+    local_files_only=True  # ensure only local files are used
 )
 
+# Define a simple function to interact with the model
 def chat(prompt: str) -> str:
     output = llm(prompt)
     return output["choices"][0]["text"]
 
-# Create a FastAPI app instance.
+# Set up FastAPI
 app = FastAPI(title="MySQL LLM Chat API")
 
-# Define Pydantic models for the API request and response.
+# Define request and response models using Pydantic
 class ChatRequest(BaseModel):
     prompt: str
 
@@ -94,27 +86,13 @@ class ChatResponse(BaseModel):
     response: str
 
 @app.post("/chat", response_model=ChatResponse)
-async def chat_endpoint(request: ChatRequest):
-    """
-    API endpoint that receives a prompt and returns the model's response.
-    The model inference is offloaded to a separate thread to avoid blocking.
-    """
-    result = await asyncio.to_thread(chat, request.prompt)
-    return ChatResponse(response=result)
-
-# Create a Gradio Interface for a user-friendly web UI.
-iface = gr.Interface(
-    fn=chat,
-    inputs="text",
-    outputs="text",
-    title="MySQL LLM Chat",
-    description="Chat interface for the MySQL finetuned model"
-)
+async def chat_endpoint(chat_request: ChatRequest):
+    prompt = chat_request.prompt
+    response_text = chat(prompt)
+    return ChatResponse(response=response_text)
 
-# Mount the Gradio app at the "/gradio" path.
-app = gr.mount_gradio_app(app, iface, path="/gradio")
-
-# Run the server with Uvicorn if this file is executed as the main module.
+# If running locally, use: uvicorn app:app --reload
 if __name__ == "__main__":
     import uvicorn
     uvicorn.run(app, host="0.0.0.0", port=8000)
+