MI

MiMo V2.5

MimoGGUF by bartowskiUpdated 2026-05-08
Download on Hugging Face
Parameters310B
Context1024K
Layers51
Architecturemimo2
Smallest quant0.5 GB
Quick answer

MiMo V2.5 needs about 206.4 GB of memory at Q4_K_M (a 188.8 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 27 quants, from 0.5 GB to 329.3 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 68.5 GB.

MiMo V2.5 – 310B native omnimodal model with vision, audio, video, and text support

MiMo V2.5 is a 310B parameter Mixture of Experts model with a 1M token context window. It supports text, image, video, and audio modalities through a 729M‑parameter Vision Transformer and a dedicated audio encoder. The model is licensed under MIT.

Good at

  • Omnimodal perception
  • Long‑context reasoning
  • Agentic workflows

Limits

  • Requires large VRAM for higher quants
  • No explicit vision‑only mode
  • License is MIT

Best for

  • Multimodal content analysis
  • Large‑scale agentic tasks
  • Long‑form reasoning

Who it’s for. Developers building multimodal AI applications that need text, image, video, and audio understanding in a single model. Researchers interested in Mixture of Experts architectures and long‑context capabilities.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.5 GB 18.1 GB
Too big Get file →
IQ1_S64.5 GB (2 parts) 82.1 GB
Too big Get file →
IQ1_M71.9 GB (2 parts) 89.6 GB
Too big Get file →
IQ2_XXS83.6 GB (3 parts) 101.2 GB
Too big Get file →
IQ2_XS93.1 GB (3 parts) 110.7 GB
Too big Get file →
IQ2_S94.7 GB (3 parts) 112.3 GB
Too big Get file →
IQ2_M104.5 GB (3 parts) 122.1 GB
Too big Get file →
Q2_K108.9 GB (3 parts) 126.6 GB
Too big Get file →
Q2_K_L109.5 GB (3 parts) 127.2 GB
Too big Get file →
IQ3_XXS130.3 GB (4 parts) 147.9 GB
Too big Get file →
Q3_K_S135.1 GB (4 parts) 152.7 GB
Too big Get file →
IQ3_XS142.2 GB (4 parts) 159.8 GB
Too big Get file →
Q3_K_M142.2 GB (4 parts) 159.9 GB
Too big Get file →
Q3_K_L147.9 GB (4 parts) 165.5 GB
Too big Get file →
IQ3_M148.3 GB (4 parts) 166.0 GB
Too big Get file →
Q3_K_XL148.5 GB (4 parts) 166.1 GB
Too big Get file →
IQ4_XS165.7 GB (5 parts) 183.3 GB
Too big Get file →
IQ4_NL175.3 GB (5 parts) 192.9 GB
Too big Get file →
Q4_0175.8 GB (5 parts) 193.4 GB
Too big Get file →
Q4_K_S181.9 GB (5 parts) 199.5 GB
Too big Get file →
Q4_K_M188.8 GB (5 parts) 206.4 GB
Too big Get file →
Q4_K_L189.2 GB (5 parts) 206.8 GB
Too big Get file →
Q4_1194.3 GB (5 parts) 211.9 GB
Too big Get file →
Q5_K_S214.3 GB (6 parts) 231.9 GB
Too big Get file →
Q5_K_M221.4 GB (6 parts) 239.1 GB
Too big Get file →
Q6_K267.6 GB (7 parts) 285.2 GB
Too big Get file →
Q8_0329.3 GB (9 parts) 346.9 GB
Too big Get file →

For 8 GB VRAM, use IQ1_S (64.52 GB file, 82.1 GB memory at 8 k context). For 16 GB VRAM, IQ2_XS (93.07 GB file, 110.7 GB memory) is suitable. For 24 GB VRAM, Q4_K_M (188.77 GB file, 206.4 GB memory) provides a balance of quality and memory usage. Estimates are approximate.

What GPU do you need for MiMo V2.5?

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: 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: the best quant that fits is the main file (0.5 GB file, about 18.1 GB in use).
  • 32 GB VRAM: the best quant that fits is the main file (0.5 GB file, about 18.1 GB in use).
  • 48 GB VRAM: the best quant that fits is the main file (0.5 GB file, about 18.1 GB in use).

Which MiMo V2.5 quant should you choose?

Q4_K_M is the usual starting point for MiMo V2.5: it balances file size and quality well. Here it is a 188.8 GB file.

The Q8_0 file is 329.3 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

Architecturemimo2
Parameters310B (approximate, mixture of experts)
Layers51
Trained context1,048,576 tokens
KV cache (fp16)2040 KB per token

Source

Quantized bybartowski
FamilyMimo
Licensemit
Downloads (HF)1,341
Last updated2026-05-08

Frequently asked questions about MiMo V2.5

How much VRAM does MiMo V2.5 need?

About 206.4 GB at Q4_K_M with an 8K context window. That is the 188.8 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 MiMo V2.5 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 18.1 GB, so you would need a larger GPU or CPU offloading.

What is the file size of MiMo V2.5 GGUF?

It depends on the quant: file is 0.5 GB, IQ1_S is 64.5 GB, IQ1_M is 71.9 GB, IQ2_XXS is 83.6 GB, IQ2_XS is 93.1 GB, IQ2_S is 94.7 GB and 21 more.

What context length does MiMo V2.5 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 17.1 GB at 8K and 68.5 GB at 32K.

Where can I download MiMo V2.5 GGUF?

The files are on Hugging Face in the bartowski/MiMo-V2.5-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 MiMo V2.5 use?

Hugging Face lists the license as mit. 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 MiMo V2.5 on 8 GB VRAM?

Only the lowest‑quantized files such as IQ1_S or IQ1_M are feasible, but they still require significant VRAM and may be slow.

Does MiMo V2.5 support vision only?

The model includes vision capabilities, but it is designed as a multimodal system and does not provide a dedicated vision‑only mode.

Other GGUFs of MiMo-V2.5

Similar models

Browse all →

More from bartowski

All by bartowski →

More Mimo models

All Mimo →

Other large models

All large models →