app.py

import discord
from discord.ext import commands
from discord.ext.commands import Context
from fastapi import FastAPI, HTTPException, Request
from fastapi.responses import JSONResponse
import gradio as gr

# Initialize FastAPI app
app = FastAPI()

# Initialize Gradio Llama model
llm = gr.Llama(model_path="model.gguf", n_ctx=4000, n_threads=2, chat_format="chatml")

# Initialize Discord bot
bot = commands.Bot(command_prefix='&')  # Define the command prefix

# Global variable to store the channel where chats will be sent
chat_channel = None

# Define the command to set the chat channel
@bot.command()
async def set_channel(ctx: Context, channel: discord.TextChannel):
    global chat_channel
    chat_channel = channel
    await ctx.send(f"Chat channel set to {channel.mention}")

# Define the function to handle the chat endpoint
@app.post("/api/v1/chat")
async def chat_post(request: Request):
    global chat_channel
    if chat_channel is None:
        raise HTTPException(status_code=400, detail="Chat channel is not set")

    data = await request.json()
    message = data.get("message")
    history = data.get("history", [])
    temperature = data.get("temperature", 0.3)
    max_tokens = data.get("max_tokens", 512)

    async def generate():
        system_prompt = "You are OpenChat, a useful AI assistant."
        formatted_prompt = [{"role": "system", "content": system_prompt}]
        for user_prompt, bot_response in history:
            formatted_prompt.append({"role": "user", "content": user_prompt})
            formatted_prompt.append({"role": "assistant", "content": bot_response})
        formatted_prompt.append({"role": "user", "content": message})
        stream_response = llm.create_chat_completion(messages=formatted_prompt, temperature=temperature,
                                                       max_tokens=max_tokens, stream=True)
        response = ""
        for chunk in stream_response:
            if len(chunk['choices'][0]["delta"]) != 0 and "content" in chunk['choices'][0]["delta"]:
                response += chunk['choices'][0]["delta"]["content"]
            yield response

    # Send the generated response to the chat channel
    async for response in generate():
        await chat_channel.send(response)

    return JSONResponse(content={"response": "Message sent to chat channel"})

# Define the function to handle the GET request for chat
@app.get("/api/v1/chat")
async def chat_get():
    return {"message": "Send a POST request to this endpoint to chat."}

# Run the bot
if __name__ == "__main__":
    import uvicorn
    uvicorn.run(app, host="0.0.0.0", port=8000)
    bot.run('YOUR_DISCORD_BOT_TOKEN')  # Replace 'YOUR_DISCORD_BOT_TOKEN' with your actual bot token