--- title: MiniMax-M3.1-Flash-Quark-MXFP4 canonical_url: "https://www.modelscope.cn/models/amd/MiniMax-M3.1-Flash-Quark-MXFP4" md_url: "https://www.modelscope.cn/models/amd/MiniMax-M3.1-Flash-Quark-MXFP4.md" repository: amd/MiniMax-M3.1-Flash-Quark-MXFP4 last_updated: 2026-10-26 license: other base_model: - MiniMaxAI/MiniMax-M3.1-Flash base_model_relation: finetune parameters: 220.5B tensor_type: - U8 - F8_E4M3 - BF16 - F32 library_name: - safetensors - pytorch frameworks: - pytorch downloads: 20 stars: 1 --- # MiniMax-M3.1-Flash-Quark-MXFP4 > MiniMax-M3.1-Flash-Quark-MXFP4 - amd 在 ModelScope 开源的模型。Model Architecture: MiniMaxM3SparseForConditionalGeneration Input: Text, Image, Video Output: Text Supported Hardware Microarchitecture: AMD MI350/MI355 ROCm: 7.2.3 PyTorch: 2.13.0 Transformers: 5.19.0… amd/MiniMax-M3.1-Flash-Quark-MXFP4 是 ModelScope 魔搭社区上的 220.5B 参数机器学习模型，采用 other 许可，基于 MiniMaxAI/MiniMax-M3.1-Flash 构建。 - **Repository**: amd/MiniMax-M3.1-Flash-Quark-MXFP4 - **License**: other - **Parameters**: 220.5B - **Base model**: MiniMaxAI/MiniMax-M3.1-Flash - **Downloads**: 20 - **Stars**: 1 - **Last updated**: 2026-10-26 Source: https://www.modelscope.cn/models/amd/MiniMax-M3.1-Flash-Quark-MXFP4 --- # Model Overview - **Model Architecture:** MiniMaxM3SparseForConditionalGeneration - **Input:** Text, Image, Video - **Output:** Text - **Supported Hardware Microarchitecture:** AMD MI350/MI355 - **ROCm**: 7.2.3 - **PyTorch**: 2.13.0 - **Transformers**: 5.19.0 - **Operating System(s):** Linux - **Inference Engine:** vLLM, SGLang - **Model Optimizer:** [AMD-Quark](https://quark.docs.amd.com/latest/index.html) - **Routed-expert weight quantization:** OCP MXFP4, Static - **Routed-expert activation quantization:** OCP MXFP4, Dynamic - **All other layers**: unchanged precision compared to base [MiniMaxAI/MiniMax-M3.1-Flash](https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash) # Model serving Please refer to [vLLM recipe](https://recipes.vllm.ai/) and [SGLang cookbook](https://docs.sglang.io/cookbook/autoregressive/MiniMax/MiniMax-M3.1). Example usage: ## vLLM ```bash VLLM_ROCM_USE_AITER=1 \ vllm serve amd/MiniMax-M3.1-Flash-Quark-MXFP4 \ --tensor-parallel-size 4 \ --reasoning-parser minimax_m3 \ --block-size 128 \ --speculative-config '{"method":"dspark","model":"amd/MiniMax-M3.1-Flash-Quark-MXFP4/dspark","num_speculative_tokens":7,"attention_backend":"FLASHINFER_TS","kv_cache_dtype":"fp8","draft_sample_method":"probabilistic","rejection_sample_method":"block"}' \ --disable-hybrid-kv-cache-manager" ``` ## SGLang ```bash SGLANG_USE_AITER=1 \ sglang serve \ --trust-remote-code \ --model-path amd/MiniMax-M3.1-Flash-Quark-MXFP4 \ --reasoning-parser minimax-m3 \ --tool-call-parser minimax-m3 \ --tp 4 \ --speculative-algorithm DSPARK \ --speculative-use-block-verification \ --cuda-graph-backend-prefill breakable \ --fp8-gemm-backend aiter \ --chunked-prefill-size 32768 \ --max-running-requests 128 \ --cuda-graph-max-bs-decode 128 \ --mem-fraction-static 0.8 \ --tokenizer-worker-num 8 \ --host 0.0.0.0 \ --port 30000 ``` # Model Quantization The model's routed experts were quantized from NVFP4 to MXFP4 using [AMD-Quark](https://quark.docs.amd.com/latest/index.html). **Quantization script:** ```bash python quantize_quark.py \ --model_dir MiniMaxAI/MiniMax-M3.1-Flash \ --quant_scheme mxfp4 \ --output_dir amd/MiniMax-M3.1-Flash-Quark-MXFP4 \ --skip_evaluation --model_export hf_format --file2file_quantization \ --exclude_layers \ '*lm_head' \ '*vision_tower*' \ '*multi_modal_projector*' \ '*patch_merge_mlp*' \ '*self_attn*' \ '*block_sparse_moe.gate' \ 'language_model.model.layers.[0-2].mlp.*' \ '*block_sparse_moe.shared_experts.*' \ '*dspark*' \ --keep_excluded_layers_as_original_model_state ``` # Evaluation The model was evaluated on GSM8K using vLLM and EvalScope. ### Accuracy
Benchmark MiniMax/MiniMax-M3.1-Flash amd/MiniMax-M3.1-Flash-Quark-MXFP4 (this model) Recovery
GSM8K (EvalScope, 4-shot) 97.2% 96.9% 99.69%
### Reproduction The GSM8K results were obtained using EvalScope and vLLM. #### Launching server ```bash VLLM_ROCM_USE_AITER=1 VLLM_ROCM_USE_AITER_MOE=1 HIP_VISIBLE_DEVICES=0,1,2,3 \ vllm serve amd/MiniMax-M3.1-Flash-Quark-MXFP4 \ --tensor-parallel-size 4 \ --host 0.0.0.0 \ --port 8101 \ --trust-remote-code \ --gpu-memory-utilization 0.9 \ --max-model-len 131072 \ --max-num-seqs 80 \ --max-num-batched-tokens 32768 \ --kv-cache-dtype fp8 \ --reasoning-parser minimax_m3 \ --moe-backend aiter ``` #### Evaluating model in a new terminal ```bash evalscope eval \ --model amd/MiniMax-M3.1-Flash-Quark-MXFP4 \ --api-url http://localhost:8101/v1/chat/completions \ --api-key EMPTY \ --datasets gsm8k \ --dataset-hub huggingface \ --dataset-args '{"gsm8k":{"dataset_id":"openai/gsm8k"}}' \ --dataset-dir hf_hub_cache \ --eval-batch-size 32 \ --generation-config '{"temperature":1,"top_p":1,"max_tokens":8192,"timeout":60000,"extra_body":{"chat_template_kwargs":{"thinking_mode":"enabled"}}}' \ --work-dir logs/gsm8k ``` # Others ## Experimental MXFP4 shared experts MXFP4 [online](https://docs.vllm.ai/en/stable/features/quantization/online/) shared experts quantization and shared experts fusion can optionally be enabled: ```bash VLLM_ROCM_USE_AITER=1 VLLM_ROCM_USE_AITER_FUSION_SHARED_EXPERTS=1 vllm serve \ amd/MiniMax-M3.1-Flash-Quark-MXFP4 \ ... \ --quantization-config.targets '{"*shared_expert*": "mxfp4"}' ``` but is experimental and not the recommended default setup. # License Modifications Copyright(c) 2026 Advanced Micro Devices, Inc. All rights reserved.