Modelpublic

aws-prototyping/Qwen3-Coder-480B-A35B-Instruct_MXFP4

sourceHugging Faceapache-2.0updated 9mo agoView on Hugging Face

1likes20downloads

Model Card

Qwen3-Coder-480B-A35B-Instruct_MXFP4

This checkpoint is a variant of Qwen3-Coder-480B-A35B-Instruct, where expert weights have been quantized to MXFP4 format similarly to gpt-oss-20b and gpt-oss-120b.

For quantizing weights we used the function downcast_to_mxfp from triton-kernels.

The checkpoint might come with a small drop in accuracy, but has ~72% size reduction compared to the original BF16 checkpoint.

Model	GSM8K (strict-match)	GSM8K (flexible-extract)
Qwen3-Coder-480B-A35B-Instruct (BF16)	89.16% ± 0.86%	90.52% ± 0.81%
Qwen3-Coder-480B-A35B-Instruct_MXFP4	89.99% ± 0.83%	90.75% ± 0.80%

Model	Size	Reduction
Qwen3-Coder-480B-A35B-Instruct (BF16)	895 GB	-
Qwen3-Coder-480B-A35B-Instruct_MXFP4	255 GB	72% smaller