vision / app.py
abdullahalioo's picture
Update app.py
e7b1f60 verified
raw
history blame
2.07 kB
import os
from fastapi import FastAPI, HTTPException, Query
from fastapi.responses import StreamingResponse
from openai import AsyncOpenAI
app = FastAPI()
# Define available models (you can expand this list)
AVAILABLE_MODELS = {
"openai/gpt-4.1-mini": "OpenAI GPT-4 Mini",
"deepseek/DeepSeek-V3-0324": "DeepSeek V3 0324",
# Add more models as needed
}
async def generate_ai_response(prompt: str, model: str):
# Configuration for unofficial GitHub AI endpoint
token = os.getenv("GITHUB_TOKEN")
if not token:
raise HTTPException(status_code=500, detail="GitHub token not configured")
endpoint = "https://models.github.ai/inference"
# Validate the model
if model not in AVAILABLE_MODELS:
raise HTTPException(status_code=400, detail=f"Model not available. Choose from: {', '.join(AVAILABLE_MODELS.keys())}")
client = AsyncOpenAI(base_url=endpoint, api_key=token)
try:
stream = await client.chat.completions.create(
messages=[
{"role": "system", "content": "You are a helpful assistant named Orion and made by Abdullah Ali"},
{"role": "user", "content": prompt}
],
model=model,
temperature=1.0,
top_p=1.0,
stream=True
)
async for chunk in stream:
if chunk.choices and chunk.choices[0].delta.content:
yield chunk.choices[0].delta.content
except Exception as err:
yield f"Error: {str(err)}"
raise HTTPException(status_code=500, detail="AI generation failed")
@app.post("/generate")
async def generate_response(
prompt: str = Query(..., description="The prompt for the AI"),
model: str = Query("openai/gpt-4.1-mini", description="The model to use for generation")
):
if not prompt:
raise HTTPException(status_code=400, detail="Prompt cannot be empty")
return StreamingResponse(
generate_ai_response(prompt, model),
media_type="text/event-stream"
)
def get_app():
return app