GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
intel-ai
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
intel-ai
Model Tree
google/gemma-4-31B
Input Modalities
Output Modalities
Supported Functionality
This model is an MXFP8 (W8A8, group_size 32) model of google/gemma-4-31B generated by intel/auto-round using RTN. Please follow the license of the original model.
bash
uv pip install vllm==0.32.0bash
vllm serve intel-ai/gemma-4-31B-MXFP8-FP8KV-FP8Attn-CT-AutoRound \ --attention-backend TRITON_ATTN \ --linear-backend marlin \ --language-model-only \ --dtype bfloat16 \ --kv-cache-dtype fp8| Benchmark | BF16 | MXFP8 | Recovery |
|---|---|---|---|
| PIQA | 82.81 | 82.48 | 99.61% |
| HellaSwag | 65.96 | 65.73 | 99.65% |
| MMLU | 81.17 | 81.16 | 99.99% |
| GSM8K | 86.28 | 85.67 | 99.30% |
| Average | 79.05 | 78.76 | 99.63% |
bash
auto-round \ --model_name google/gemma-4-31B \ --scheme MXFP8 \ --iters 0 \ --nsamples 128 \ --static_kv_dtype fp8 \ --static_attention_dtype fp8 \ --format llm_compressor \ --output_dir gemma-4-31B-MXFP8-FP8KV-FP8Attn-CT-AutoRoundThe model can produce factually incorrect output, and should not be relied on to produce factually accurate information. Because of the limitations of the pretrained model and the finetuning datasets, it is possible that this model could generate lewd, biased or otherwise offensive outputs. Therefore, before deploying any applications of the model, developers should perform safety testing.
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Here are a couple of useful links to learn more about Intel's AI software:
The license on this model does not constitute legal advice. We are not responsible for the actions of third parties who use this model. Please consult an attorney before using this model for commercial purposes.
text
@article{cheng2023optimize, title={Optimize weight rounding via signed gradient descent for the quantization of llms}, author={Cheng, Wenhua and Zhang, Weiwei and Shen, Haihao and Cai, Yiyang and He, Xin and Lv, Kaokao and Liu, Yi}, journal={arXiv preprint arXiv:2309.05516}, year={2023}}Signed-off-by: Sun, Xuehao xuehao.sun@intel.com
vllm serve intel-ai/gemma-4-31B-MXFP8-FP8KV-FP8Attn-CT-AutoRound \ --attention-backend TRITON_ATTN \ --linear-backend marlin \ --language-model-only \ --dtype bfloat16 \ --kv-cache-dtype fp8auto-round \ --model_name google/gemma-4-31B \ --scheme MXFP8 \ --iters 0 \ --nsamples 128 \ --static_kv_dtype fp8 \ --static_attention_dtype fp8 \ --format llm_compressor \ --output_dir gemma-4-31B-MXFP8-FP8KV-FP8Attn-CT-AutoRound@article{cheng2023optimize, title={Optimize weight rounding via signed gradient descent for the quantization of llms}, author={Cheng, Wenhua and Zhang, Weiwei and Shen, Haihao and Cai, Yiyang and He, Xin and Lv, Kaokao and Liu, Yi}, journal={arXiv preprint arXiv:2309.05516}, year={2023}}