MI

MiniMax M3

MinimaxGGUF by bartowskiUpdated 2026-07-30
Download on Hugging Face
Parameters426B
Context1024K
Layers60
Architectureminimax-m3
Smallest quant0.5 GB
Quick answer

MiniMax M3 needs about 262.8 GB of memory at Q4_K_M (a 261.3 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 26 quants, from 0.5 GB to 453.6 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 4.0 GB.

MiniMax M3 – a 426B multimodal LLM with 1M token context

MiniMax M3 is a native multimodal model with 1M context and 426B parameters. It uses MiniMax Sparse Attention for efficient long‑context processing. The model supports text, image, and video inputs and offers reasoning modes via the `thinking` parameter.

Good at

  • Long‑context reasoning
  • Multimodal input handling
  • Coding and cowork tasks

Limits

  • Requires large VRAM for full precision
  • No vision support without mmproj file
  • License is other, not standard

Best for

  • Developers building multimodal assistants
  • Researchers needing 1M token context
  • Teams working on coding and cowork workflows

Who it’s for. The model is suitable for developers and researchers who need a multimodal LLM with extensive context. It is ideal for applications that require deep semantic fusion across text, image, and video, and for tasks that benefit from long‑horizon reasoning.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.5 GB 2.0 GB
Fits Get file →
IQ1_S90.5 GB (3 parts) 92.0 GB
Too big Get file →
IQ1_M100.7 GB (3 parts) 102.2 GB
Too big Get file →
IQ2_XXS116.6 GB (3 parts) 118.1 GB
Too big Get file →
IQ2_XS129.5 GB (4 parts) 131.0 GB
Too big Get file →
IQ2_S132.1 GB (4 parts) 133.6 GB
Too big Get file →
IQ2_M145.6 GB (4 parts) 147.1 GB
Too big Get file →
Q2_K151.9 GB (4 parts) 153.4 GB
Too big Get file →
Q2_K_L153.1 GB (4 parts) 154.6 GB
Too big Get file →
IQ3_XXS180.0 GB (5 parts) 181.5 GB
Too big Get file →
Q3_K_S187.3 GB (5 parts) 188.8 GB
Too big Get file →
IQ3_XS196.4 GB (5 parts) 197.9 GB
Too big Get file →
Q3_K_M196.6 GB (5 parts) 198.1 GB
Too big Get file →
Q3_K_L205.0 GB (6 parts) 206.5 GB
Too big Get file →
IQ3_M205.5 GB (6 parts) 207.0 GB
Too big Get file →
Q3_K_XL206.0 GB (6 parts) 207.6 GB
Too big Get file →
IQ4_XS229.7 GB (6 parts) 231.2 GB
Too big Get file →
IQ4_NL242.8 GB (7 parts) 244.3 GB
Too big Get file →
Q4_0243.6 GB (7 parts) 245.2 GB
Too big Get file →
Q4_K_S251.4 GB (7 parts) 252.9 GB
Too big Get file →
Q4_K_M261.3 GB (7 parts) 262.8 GB
Too big Get file →
Q4_1268.9 GB (7 parts) 270.4 GB
Too big Get file →
Q5_K_S295.2 GB (8 parts) 296.7 GB
Too big Get file →
Q5_K_M305.3 GB (8 parts) 306.9 GB
Too big Get file →
Q6_K369.4 GB (10 parts) 370.9 GB
Too big Get file →
Q8_0453.6 GB (12 parts) 455.1 GB
Too big Get file →

For 8‑GB VRAM, use Q4_K_M (261.28 GB) or smaller quants. For 16‑GB VRAM, Q4_K_L (253.09 GB) or Q5_K_S (295.23 GB) are suitable. For 24‑GB VRAM, Q5_K_M (305.35 GB) or Q6_K (369.40 GB) provide higher quality. Estimates are approximate.

What GPU do you need for MiniMax M3?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is the main file (0.5 GB file, about 2.0 GB in use).
  • 6 GB VRAM: the best quant that fits is the main file (0.5 GB file, about 2.0 GB in use).
  • 8 GB VRAM: the best quant that fits is the main file (0.5 GB file, about 2.0 GB in use).
  • 12 GB VRAM: the best quant that fits is the main file (0.5 GB file, about 2.0 GB in use).
  • 16 GB VRAM: the best quant that fits is the main file (0.5 GB file, about 2.0 GB in use).
  • 24 GB VRAM: the best quant that fits is the main file (0.5 GB file, about 2.0 GB in use).
  • 32 GB VRAM: the best quant that fits is the main file (0.5 GB file, about 2.0 GB in use).
  • 48 GB VRAM: the best quant that fits is the main file (0.5 GB file, about 2.0 GB in use).

Which MiniMax M3 quant should you choose?

Q4_K_M is the usual starting point for MiniMax M3: it balances file size and quality well. Here it is a 261.3 GB file.

The Q8_0 file is 453.6 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is the first one in the table at 0.5 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureminimax-m3
Parameters426B (approximate, mixture of experts)
Layers60
Trained context1,048,576 tokens
KV cache (fp16)120 KB per token

Source

Quantized bybartowski
FamilyMinimax
Licenseother
Downloads (HF)2,723
Last updated2026-07-30

Frequently asked questions about MiniMax M3

How much VRAM does MiniMax M3 need?

About 262.8 GB at Q4_K_M with an 8K context window. That is the 261.3 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run MiniMax M3 on an 8 GB GPU?

Yes, with the smaller files: the 0.5 GB file needs about 2.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of MiniMax M3 GGUF?

It depends on the quant: file is 0.5 GB, IQ1_S is 90.5 GB, IQ1_M is 100.7 GB, IQ2_XXS is 116.6 GB, IQ2_XS is 129.5 GB, IQ2_S is 132.1 GB and 20 more.

What context length does MiniMax M3 support?

The GGUF header lists a trained context length of 1,048,576 tokens (1024K). The KV cache grows with the context you use: about 1.0 GB at 8K and 4.0 GB at 32K.

Where can I download MiniMax M3 GGUF?

The files are on Hugging Face in the bartowski/MiniMax-M3-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does MiniMax M3 use?

Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run MiniMax M3 on 8 GB VRAM?

Only the smallest quants such as Q4_K_M or smaller can fit, but performance may be limited.

Does the model support vision without extra files?

Vision requires the mmproj file; the base model does not include it.

What inference frameworks support this model?

Supported frameworks include llama.cpp, vLLM, Transformers, KTransformers, unsloth, and ATOM.

Other GGUFs of MiniMax-M3

Similar models

Browse all →

More from bartowski

All by bartowski →

More Minimax models

All Minimax →

Other large models

All large models →