openquant-args.json

{"model": "meta-llama/Llama-3.3-70B-Instruct", "weight_block_size": null}