Skip to content

Fix critical quantization parameter bugs in AWQ and NVFP4 #6

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
j4ys0n merged 1 commit into from
Nov 4, 2025
Merged

Fix critical quantization parameter bugs in AWQ and NVFP4 #6

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 7 additions & 7 deletions src/msquant/core/quantizer/engine.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -108,7 +108,7 @@ def build_recipe(config: QuantizationConfig):
"weights": {
"num_bits": config.w_bit,
"type": "int",
"symmetric": config.zero_point,
"symmetric": not config.zero_point,
"strategy": "group",
"group_size": config.group_size,
},
Expand Down Expand Up @@ -146,10 +146,10 @@ def run(config: QuantizationConfig, logger: QuantizationLogger):

if config.calib_config:
oneshot_kwargs["dataset_config_name"] = config.calib_config

if config.calib_split:
oneshot_kwargs["dataset_split"] = config.calib_split
oneshot_kwargs["splits"] = {"calibration": config.calib_split}

logger.info("Starting LLM-Compressor oneshot() with AWQ", {
"model": oneshot_kwargs["model"],
"dataset": oneshot_kwargs["dataset"],
Expand Down Expand Up @@ -220,10 +220,10 @@ def run(config: QuantizationConfig, logger: QuantizationLogger):

if config.calib_config:
oneshot_kwargs["dataset_config_name"] = config.calib_config

if config.calib_split:
oneshot_kwargs["dataset_split"] = config.calib_split
oneshot_kwargs["splits"] = {"calibration": config.calib_split}

logger.info("Starting LLM-Compressor oneshot() with FP4 quantization", {
"model": oneshot_kwargs["model"],
"dataset": oneshot_kwargs["dataset"],
Expand Down