MiniMax M2.5 NVFP4 is an open-source language model by nvidia. Features: 116.3b LLM, VRAM: 140.1GB, Context: 192K, License: other, LLM Explorer Score: 0.36.
8-bit Base model:minimaxai/minimax-m... Base model:quantized:minimaxai... Conversational Custom code Minimax Minimax m2 Model optimizer Modelopt Nvfp4 Nvidia Quantized Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/nvidia/MiniMax-M2.5-NVFP4
MiniMax M2.5 NVFP4 Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
MiniMax M2.5 NVFP4 Parameters and Internals
| LLM Name | MiniMax M2.5 NVFP4 |
| Repository 🤗 | https://huggingface.co/nvidia/MiniMax-M2.5-NVFP4 |
| Base Model(s) | |
| Model Size | 116.3b |
| Required VRAM | 140.1 GB |
| Updated | 2026-03-31 |
| Maintainer | nvidia |
| Model Type | minimax_m2 |
| Model Files | |
| Model Architecture | MiniMaxM2ForCausalLM |
| License | other |
| Context Length | 196608 |
| Model Max Length | 196608 |
| Transformers Version | 4.57.1 |
| Tokenizer Class | GPT2Tokenizer |
| Vocabulary Size | 200064 |
Rank the MiniMax M2.5 NVFP4 Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 52392 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a