app.py

import subprocess
from huggingface_hub import hf_hub_download

# 1. Download the model
repo_id = "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF"
filename = "Meta-Llama-3.1-8B-Instruct-Q4_K_M.gguf"

print("Downloading model for OpenClaw...")
model_path = hf_hub_download(repo_id=repo_id, filename=filename)

# 2. Launch the OpenAI-compatible server
print("Starting server on port 7860...")
subprocess.run([
    "python", "-m", "llama_cpp.server",
    "--model", model_path,
    "--host", "0.0.0.0",
    "--port", "7860",
    "--n_ctx", "2048",
    "--n_threads", "2"
])