Model Overview
Description:
The NVIDIA MiniMax-M2.5-NVFP4 model is the quantized version of MiniMax's MiniMax-M2.5 model, which is an auto-regressive language model that uses an optimized transformer architecture. For more information, please check here. The NVIDIA MiniMax-M2.5 NVFP4 model is quantized with Nvidia Model Optimizer.
This model is ready for commercial/non-commercial use.
Third-Party Community Consideration
This model is not owned or developed by NVIDIA. This model has been developed and built to a third-party's requirements for this application and use case; see link to Non-NVIDIA (MiniMax-M2.5) Model Card.
License/Terms of Use:
Governing Terms: Use of this model is governed by the NVIDIA Open Model License .
ADDITIONAL INFORMATION : Modified MIT License. MiniMax M2.5 .
Deployment Geography:
Global
Use Case:
Developers looking to take off-the-shelf, pre-quantized models for deployment in AI Agent systems, chatbots, RAG systems, and other AI-powered applications.
Release Date:
Huggingface 03/25/2026 via https://huggingface.co/nvidia/MiniMax-M2.5-NVFP4
Model Architecture:
Architecture Type: Transformers Network Architecture: MiniMaxM2ForCausalLM **This model was developed based on MiniMax-M2.5 **Number of model parameters: 229B
Input:
Input Type(s): Text Input Format(s): String Input Parameters: 1D (One-Dimensional): Sequences Other Properties Related to Input: Context length 196,608 tokens
Output:
Output Type(s): Text Output Format: String Output Parameters: 1D (One-Dimensional): Sequences Other Properties Related to Output: None
Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated systems. By leveraging NVIDIA's hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions.
Software Integration:
Runtime Engine(s):
- SGLang
- vLLM
Supported Hardware Microarchitecture Compatibility:
- NVIDIA Blackwell
Preferred Operating System(s):
- Linux
The integration of foundation and fine-tuned models into AI systems requires additional testing using use-case-specific data to ensure safe and effective deployment. Following the V-model methodology, iterative testing and validation at both unit and system levels are essential to mitigate risks, meet technical and functional requirements, and ensure compliance with safety and ethical standards before deployment.
Model Version(s):
The model is quantized with nvidia-modelopt v0.43.0
Training and Evaluation Datasets:
Calibration Dataset:
** Link: cnn_dailymail, Nemotron-Post-Training-Dataset-v2 ** Data Collection Method by dataset: Automated. ** Labeling Method by dataset: Automated. ** Properties: The cnn_dailymail dataset is an English-language dataset containing just over 300k unique news articles as written by journalists at CNN and the Daily Mail. The Nemotron-Post-Training-Dataset-v2 is a post-training dataset curated by NVIDIA containing multi-turn conversations across diverse topics.
Training Dataset:
** Data Modality: Text ** Data Collection Method by dataset: Undisclosed ** Labeling Method by dataset: Undisclosed ** Properties: Undisclosed
Evaluation Dataset:
- Datasets: MMLU Pro, GPQA Diamond, LiveCodeBench V6, SciCode, AIME 2025, AA-LCR, IFBench ** Data Collection Method by dataset: Hybrid: Automated, Human ** Labeling Method by dataset: Hybrid: Human, Automated ** Properties: We evaluated the model on text-based reasoning and coding benchmarks: MMLU Pro is a multi-task language understanding benchmark with challenging multiple-choice questions across diverse academic domains; GPQA Diamond contains 448 graduate-level multiple-choice questions written by domain experts in biology, physics, and chemistry; LiveCodeBench V6 contains competitive programming problems; SciCode evaluates scientific coding capabilities; AIME 2025 contains problems from the American Invitational Mathematics Examination; AA-LCR (Artificial Analysis Long Context Recall) evaluates a model's ability to accurately retrieve and recall information from long input contexts; IFBench is a benchmark for evaluating instruction-following capabilities across diverse and structured task constraints.
Inference:
Engine: SGLang Test Hardware: B200
Post Training Quantization
This model was obtained by quantizing the weights and activations of MiniMax-M2.5 to NVFP4 data type, ready for inference with SGLang. Only the weights and activations of the linear operators within transformer blocks are quantized. This optimization reduces the number of bits per parameter from 8 to 4, reducing the disk size and GPU memory requirements by approximately 1.65x.
Usage
To serve this checkpoint with SGLang, you can start the docker lmsysorg/sglang:latest and run the sample command below:
python3 -m sglang.launch_server --model nvidia/MiniMax-M2.5-NVFP4 \
--tensor-parallel-size 8 \
--quantization modelopt_fp4 \
--trust-remote-code \
--reasoning-parser minimax-append-think \
--tool-call-parser minimax-m2 \
--moe-runner-backend flashinfer_cutlass \
--attention-backend flashinfer
To serve this checkpoint with vLLM, you can launch the docker image vllm/vllm-openai:latest and run the sample command below:
vllm serve nvidia/MiniMax-M2.5-NVFP4 \
--tensor-parallel-size 8 \
--tool-call-parser minimax_m2 \
--reasoning-parser minimax_m2_append_think \
--enable-auto-tool-choice \
--trust-remote-code
Evaluation
The accuracy benchmark results are presented in the table below:
Baseline: MiniMax-M2.5. Benchmarked with temperature=1.0, top_p=0.95, max num tokens 64000
Ethical Considerations
NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse.
Please report model quality, risk, security vulnerabilities or NVIDIA AI Concerns here.