amd
/

Qwen3-Coder-Next-MXFP4

8-bit precision

Model card Files Files and versions

linzhao-amd commited on 27 days ago

Commit

c4164ba

·

verified ·

1 Parent(s): 3213b47

Update README.md

Files changed (1) hide show

README.md +3 -0

README.md CHANGED Viewed

@@ -15,6 +15,9 @@ license: apache-2.0
   - **moe**
     - **Weight quantization:** MOE-only, OCP MXFP4, Static
     - **Activation quantization:** MOE-only, OCP MXFP4, Dynamic
 - **Calibration Dataset:** [Pile](https://huggingface.co/datasets/mit-han-lab/pile-val-backup)
 This model was built with Qwen3-Coder-Next model by applying AMD-Quark for MXFP4 quantization.

   - **moe**
     - **Weight quantization:** MOE-only, OCP MXFP4, Static
     - **Activation quantization:** MOE-only, OCP MXFP4, Dynamic
+  - **attn (linear_attn.out_proj/self_attn.o_proj)**
+    - **Weight quantization:** MOE-only, OCP MXFP4, Static
+    - **Activation quantization:** MOE-only, OCP MXFP4, Dynamic
 - **Calibration Dataset:** [Pile](https://huggingface.co/datasets/mit-han-lab/pile-val-backup)
 This model was built with Qwen3-Coder-Next model by applying AMD-Quark for MXFP4 quantization.