Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -27,7 +27,8 @@ python3 quantize_quark.py \
|
|
| 27 |
--num_calib_data 128 \
|
| 28 |
--model_export quark_safetensors \
|
| 29 |
--multi_gpu \
|
| 30 |
-
--no_weight_matrix_merge
|
|
|
|
| 31 |
```
|
| 32 |
## Deployment
|
| 33 |
Quark has its own export format and allows FP8 quantized models to be efficiently deployed using the vLLM backend(vLLM-compatible).
|
|
|
|
| 27 |
--num_calib_data 128 \
|
| 28 |
--model_export quark_safetensors \
|
| 29 |
--multi_gpu \
|
| 30 |
+
--no_weight_matrix_merge \
|
| 31 |
+
--custom_mode fp8
|
| 32 |
```
|
| 33 |
## Deployment
|
| 34 |
Quark has its own export format and allows FP8 quantized models to be efficiently deployed using the vLLM backend(vLLM-compatible).
|