Hugging Face's logo

amd
/
DeepSeek-R1-MXFP4-ASQ

Safetensors
deepseek_v3
custom_code
8-bit precision
quark
Model card Files
xet
Community
1
linzhao-amd commited on
Commit
d165184
·
verified ·
1 Parent(s): b9c0909

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -1
README.md CHANGED
@@ -12,9 +12,11 @@ base_model:
12
  - **Output:** Text
13
  - **Supported Hardware Microarchitecture:** AMD MI350/MI355
14
  - **ROCm**: 7.0
 
 
15
  - **Operating System(s):** Linux
16
  - **Inference Engine:** [SGLang](https://docs.sglang.ai/)
17
- - **Model Optimizer:** [AMD-Quark](https://quark.docs.amd.com/latest/index.html)
18
  - **Weight quantization:** OCP MXFP4, Static
19
  - **Activation quantization:** OCP MXFP4, Dynamic
20
  - **Calibration Dataset:** [Pile](https://huggingface.co/datasets/mit-han-lab/pile-val-backup)
 
12
  - **Output:** Text
13
  - **Supported Hardware Microarchitecture:** AMD MI350/MI355
14
  - **ROCm**: 7.0
15
+ - **PyTorch**: 2.8.0
16
+ - **Transformers**: 4.53.0
17
  - **Operating System(s):** Linux
18
  - **Inference Engine:** [SGLang](https://docs.sglang.ai/)
19
+ - **Model Optimizer:** [AMD-Quark](https://quark.docs.amd.com/latest/index.html) (V0.10)
20
  - **Weight quantization:** OCP MXFP4, Static
21
  - **Activation quantization:** OCP MXFP4, Dynamic
22
  - **Calibration Dataset:** [Pile](https://huggingface.co/datasets/mit-han-lab/pile-val-backup)