amd
/

DeepSeek-R1-MXFP4

8-bit precision

Model card Files Files and versions

linzhao-amd commited on Aug 6, 2025

Commit

ff33cb4

·

verified ·

1 Parent(s): c59c670

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -38,7 +38,7 @@ python3 quantize_quark.py --model_dir $MODEL_DIR \
                           --group_size 32 \
                           --num_calib_data 128 \
                           --exclude_layers "lm_head" \
-                          --multi_gpu \
                           --model_export hf_format \
                           --output_dir amd/DeepSeek-R1-MXFP4
 ```

                           --group_size 32 \
                           --num_calib_data 128 \
                           --exclude_layers "lm_head" \
+                          --multi_device \
                           --model_export hf_format \
                           --output_dir amd/DeepSeek-R1-MXFP4
 ```