MI

MiniMax M2.5 PRISM PRO

MinimaxGGUF by Ex0bitUpdated 2026-02-15
Download on Hugging Face
Parameters229B
Context192K
Layers—
Architectureminimax-m2
Smallest quant60.3 GB
Quick answer

MiniMax M2.5 PRISM PRO needs about 155.2 GB of memory at Q4_K_XL (a 138.1 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 7 quants, from 60.3 GB (IQ2_XXS) to 243.1 GB (Q8_0).

MiniMax M2.5 PRISM PRO is a 229B-parameter Minimax model in GGUF format, quantized by Ex0bit. Its trained context length is 192K tokens. The repository has 7 quant files, from 60.3 GB (IQ2_XXS) to 243.1 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ2_XXS60.3 GB 68.1 GB
Too big Get file →
IQ2_M78.2 GB 88.1 GB
Too big Get file →
Q3_K_XL109.3 GB 122.9 GB
Too big Get file →
Q4_K_XL138.1 GB 155.2 GB
Too big Get file →
Q6_K187.9 GB 211.0 GB
Too big Get file →
Q6_K_XL194.6 GB 218.4 GB
Too big Get file →
Q8_0243.1 GB 272.8 GB
Too big Get file →

What GPU do you need for MiniMax M2.5 PRISM PRO?

These figures add the file size, an fp16 KV cache at an 8K context window 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: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: no quant fits at an 8K context.
  • 32 GB VRAM: no quant fits at an 8K context.
  • 48 GB VRAM: no quant fits at an 8K context.

Which MiniMax M2.5 PRISM PRO quant should you choose?

Q4_K_XL (138.1 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.

If memory is tight, the smallest file is IQ2_XXS at 60.3 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureminimax-m2
Parameters229B
Trained context196,608 tokens

Source

Quantized byEx0bit
FamilyMinimax
Licenseother
Downloads (HF)0
Last updated2026-02-15

Frequently asked questions about MiniMax M2.5 PRISM PRO

How much VRAM does MiniMax M2.5 PRISM PRO need?

About 155.2 GB at Q4_K_XL with an 8K context window. That is the 138.1 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 M2.5 PRISM PRO on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 68.1 GB, so you would need a larger GPU or CPU offloading.

What is the file size of MiniMax M2.5 PRISM PRO GGUF?

It depends on the quant: IQ2_XXS is 60.3 GB, IQ2_M is 78.2 GB, Q3_K_XL is 109.3 GB, Q4_K_XL is 138.1 GB, Q6_K is 187.9 GB, Q6_K_XL is 194.6 GB and 1 more.

What context length does MiniMax M2.5 PRISM PRO support?

The GGUF header lists a trained context length of 196,608 tokens (192K).

Where can I download MiniMax M2.5 PRISM PRO GGUF?

The files are on Hugging Face in the Ex0bit/MiniMax-M2.5-PRISM-PRO 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 M2.5 PRISM PRO 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.

Other GGUFs of MiniMax-M2.5

Similar models

Browse all →

More from Ex0bit

All by Ex0bit →

More Minimax models

All Minimax →

Other large models

All large models →