File size: 1,103 Bytes
d5743c4 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 | """Convert HF checkpoint -> GGUF via llama.cpp. Needs: git clone https://github.com/ggerganov/llama.cpp"""
import subprocess, sys
from pathlib import Path
def main(ckpt="outputs/tiny-50M", quant="Q4_K_M", out="outputs/tiny-50M.gguf"):
conv = Path("third_party/llama.cpp/convert_hf_to_gguf.py")
if not conv.exists():
print("Clone llama.cpp first: git clone https://github.com/ggerganov/llama.cpp third_party/llama.cpp")
sys.exit(1)
subprocess.run([sys.executable, str(conv), ckpt, "--outfile", out, "--outtype", "f16"], check=True)
quantize = Path("third_party/llama.cpp/build/bin/llama-quantize")
if quantize.exists():
subprocess.run([str(quantize), out, out.replace(".gguf", f".{quant}.gguf"), quant], check=True)
print(f"GGUF ready -> {out}")
if __name__ == "__main__":
import argparse
ap = argparse.ArgumentParser()
ap.add_argument("--ckpt", default="outputs/tiny-50M")
ap.add_argument("--quant", default="Q4_K_M")
ap.add_argument("--out", default="outputs/tiny-50M.gguf")
a = ap.parse_args()
main(a.ckpt, a.quant, a.out)
|