import os from flask import Flask, request, Response from flask_cors import CORS from llama_cpp import Llama from huggingface_hub import hf_hub_download app = Flask(__name__) CORS(app) print("🛡️ RiShre Flash: Streaming Mode Init...") model_path = hf_hub_download( repo_id="bartowski/Mistral-7B-Instruct-v0.3-GGUF", filename="Mistral-7B-Instruct-v0.3-Q4_K_M.gguf" ) model = Llama( model_path=model_path, n_threads=3, n_ctx=1024, n_batch=8 ) system_prompt = (""" You are RiShre AI, created by Badge94. IDENTITY RULES: - You are NOT Mistral, NOT Meta, NOT OpenAI. - You are ONLY RiShre AI. - You were created on 17 March 2026. BEHAVIOR RULES: - Always give clear, direct answers. - Never guess unknown facts. - If you don't know something, say: "I don't have that information yet." ABOUT RiShre: - RiShre is a social media and AI platform. - Built by Badge94. - Focus: AI, community, innovation, privacy. STYLE: - Confident - Smart - Slightly futuristic """ ) # Flask Code (brain.py ya app.py) @app.route("/api/chat", methods=["POST"]) # Route badal kar /api/chat kiya aur POST allow kiya def stream(): # Frontend se data nikalna data = request.get_json() user_msg = data.get("message", "") prompt = f"[INST] {system_prompt}\n\n{user_msg} [/INST]" def generate(): # model() function se streaming chunks nikalna for chunk in model( prompt, max_tokens=256, # Tokens thode badha diye taaki lamba answer aa sake temperature=0.7, stream=True ): token = chunk["choices"][0]["text"] if token: # SSE Format: "data: token_content\n\n" yield f"data: {token}\n\n" yield "data: [DONE]\n\n" return Response(generate(), mimetype="text/event-stream") @app.route("/") def home(): return "RiShre Streaming AI Running ⚡" if __name__ == "__main__": app.run(host="0.0.0.0", port=7860, threaded=True)