File size: 475 Bytes
11690a5 | 1 2 3 4 5 6 7 8 9 10 11 12 13 | MODEL=vila-7b
# run AWQ search (optional; we provided the pre-computed results)
python -m awq.entry --model_path /dataset/vila-hf/$MODEL \
--w_bit 4 --q_group_size 128 \
--run_awq --dump_awq awq_cache/$MODEL-w4-g128.pt
# generate real quantized weights (w4)
python -m awq.entry --model_path /dataset/vila-hf/$MODEL \
--w_bit 4 --q_group_size 128 \
--load_awq awq_cache/$MODEL-w4-g128.pt \
--q_backend real --dump_quant quant_cache/$MODEL-w4-g128-awq.pt
|