Update README.md

README.md

@@ -6,8 +6,8 @@ base_model:
 
 ## Model Details
 
-This model is an mxfp4 quantized version of [Qwen/Qwen3-235B-A22B](https://huggingface.co/Qwen/Qwen3-235B-A22B) generated by [intel/auto-round](https://github.com/intel/auto-round).
-The model is not able to be published due to the storage limitation. Please follow the INC example README to generate and evaluate the low precision model.
+This model card is for mxfp4/mxfp8 quantization of [Qwen/Qwen3-235B-A22B](https://huggingface.co/Qwen/Qwen3-235B-A22B) based on [intel/auto-round](https://github.com/intel/auto-round).
+The models are not able to be published due to the storage limitation. Please follow the INC example README to generate and evaluate the low precision models.
 
 ## How to Use
 
@@ -15,13 +15,13 @@ The step-by-step README of quantization and evaluation can be found in [Intel Ne
 
 ## Evaluate Results
 
-|   Task    | backend |  BF16  | MXFP4  |
-|:---------:|:-------:|:------:|:------:|
-| hellaswag |  vllm   | 0.6794 | 0.6680 |
-|   piqa    |  vllm   | 0.8177 | 0.8161 |
-|   mmlu    |  vllm   | 0.8492 | 0.8435 |
-|   gsm8k   |  vllm   | 0.9242 | 0.9363 |
-|  average  |  vllm   | 0.8176 | 0.8160 |
+|    Task     | backend |    BF16    |   MXFP4    |   MXFP8    |
+|:-----------:|:-------:|:----------:|:----------:|:----------:|
+|  hellaswag  |  vllm   |   0.6794   |   0.6680   |   0.6768   |
+|    piqa     |  vllm   |   0.8177   |   0.8161   |   0.8221   |
+|    mmlu     |  vllm   |   0.8492   |   0.8435   |   0.8472   |
+|    gsm8k    |  vllm   |   0.9242   |   0.9363   |   0.9325   |
+| **average** |  vllm   | **0.8176** | **0.8160** | **0.8196** |
 
 ## Ethical Considerations and Limitations