import { pipeline } from '@huggingface/transformers'; import http from 'http'; // Configuration const PORT = 7860; // UPGRADED: Starcoder2-1b is a much more capable coding model const MODEL_NAME = 'onnx-community/starcoder2-3b'; let generator; // Initialize the model on startup async function loadModel() { console.log("Loading upgraded coding model..."); // We specify 'auto' for the device to let the server manage memory efficiently generator = await pipeline('text-generation', MODEL_NAME, { device: 'auto' }); console.log("Model loaded successfully!"); } const server = http.createServer(async (req, res) => { res.setHeader('Access-Control-Allow-Origin', '*'); res.setHeader('Access-Control-Allow-Methods', 'GET, POST, OPTIONS'); res.setHeader('Access-Control-Allow-Headers', 'Content-Type'); if (req.method === 'OPTIONS') { res.writeHead(200); return res.end(); } res.setHeader('Content-Type', 'application/json'); const pathname = req.url.split('?')[0]; if (pathname === '/' && req.method === 'GET') { res.writeHead(200); res.end(JSON.stringify({ "status": "Backend running with Starcoder2-1b" })); return; } if (pathname === '/generate' && req.method === 'POST') { let body = ''; req.on('data', chunk => { body += chunk.toString(); }); req.on('end', async () => { try { const { prompt } = JSON.parse(body); if (!generator) { res.writeHead(503); return res.end(JSON.stringify({ error: "Model is still loading..." })); } // UPGRADED: Increased tokens and tuned temperature for better code quality const output = await generator(prompt, { max_new_tokens: 250, temperature: 0.2, do_sample: true }); res.writeHead(200); res.end(JSON.stringify({ result: output[0].generated_text })); } catch (err) { res.writeHead(400); res.end(JSON.stringify({ error: "Invalid JSON or request" })); } }); return; } res.writeHead(404); res.end(JSON.stringify({ error: "Not Found" })); }); loadModel().then(() => { server.listen(PORT, '0.0.0.0', () => { console.log(`Server running at http://0.0.0.0:${PORT}`); }); });