MI

MiMo-V2.6-Pro-RL

MimoGGUF by ggml-orgUpdated 2026-10-02
Download on Hugging Face
Parameters1020B
Context1024K
Layers73
Architecturemimo2
Smallest quant2.1 GB
Quick answer

MiMo-V2.6-Pro-RL needs about 57.1 GB of memory at BF16 (a 7.6 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 7 quants, from 2.1 GB (Q4_0) to 554.2 GB (MXFP4). Longer context needs more: at 32K tokens the KV cache alone adds about 196.0 GB.

MiMo-V2.6-Pro-RL: 1.02 B parameter multimodal model with 1 M token context

MiMo-V2.6-Pro-RL is a 1.02 B parameter model in the MiMo family. It supports vision, audio, text, video, and long‑horizon tasks with a 1 M token context window. The model is released under the MIT license and is available in multiple GGUF quantizations.

Good at

  • Vision and audio processing
  • Long‑horizon reasoning up to 1 M tokens
  • Mixed RL training across domains

Limits

  • Requires large VRAM for full precision
  • No explicit language support list
  • Limited to GGUF format

Best for

  • Omnimodal agents needing long context
  • Self‑improvement loops with RL
  • Vision‑enabled text generation

Who it’s for. Developers building multimodal agents that require extended context and reinforcement learning capabilities. The model is suitable for research and production use where a 1 M token context is beneficial.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_02.1 GB 51.6 GB
Too big Get file →
Q8_04.1 GB 53.5 GB
Too big Get file →
Q8_04.9 GB 54.4 GB
Too big Get file →
BF167.6 GB 57.1 GB
Too big Get file →
BF169.3 GB 58.8 GB
Too big Get file →
Q2_K418.8 GB (7 parts) 468.3 GB
Too big Get file →
MXFP4554.2 GB (7 parts) 603.7 GB
Too big Get file →

For 8 GB VRAM, use Q4_0 (2.15 GB file, ~51.6 GB memory at 8 k context). For 16 GB VRAM, Q8_0 (4.05 GB file, ~53.5 GB memory). For 32 GB VRAM, BF16 (7.63 GB file, ~57.1 GB memory). Sharded Q2_K or MXFP4 are only for very large memory budgets.

What GPU do you need for MiMo-V2.6-Pro-RL?

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: 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 MiMo-V2.6-Pro-RL quant should you choose?

BF16 (7.6 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.

If memory is tight, the smallest file is Q4_0 at 2.1 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturemimo2
Parameters1020B (approximate, mixture of experts)
Layers73
Trained context1,048,576 tokens
KV cache (fp16)5840 KB per token

Source

Quantized byggml-org
FamilyMimo
Licensemit
Downloads (HF)6,157
Last updated2026-10-02

Frequently asked questions about MiMo-V2.6-Pro-RL

How much VRAM does MiMo-V2.6-Pro-RL need?

About 57.1 GB at BF16 with an 8K context window. That is the 7.6 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-V2.6-Pro-RL on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 51.6 GB, so you would need a larger GPU or CPU offloading.

What is the file size of MiMo-V2.6-Pro-RL GGUF?

It depends on the quant: Q4_0 is 2.1 GB, Q8_0 is 4.1 GB, Q8_0 is 4.9 GB, BF16 is 7.6 GB, BF16 is 9.3 GB, Q2_K is 418.8 GB and 1 more.

What context length does MiMo-V2.6-Pro-RL 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 49.0 GB at 8K and 196.0 GB at 32K.

Where can I download MiMo-V2.6-Pro-RL GGUF?

The files are on Hugging Face in the ggml-org/MiMo-V2.6-Pro-RL-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-V2.6-Pro-RL 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-V2.6-Pro-RL on 8 GB VRAM?

Use the Q4_0 quantization; it requires about 51.6 GB memory at 8 k context, which is beyond 8 GB VRAM, so it is not feasible.

Does the model support vision?

Yes, the model includes vision and audio encoders and supports multimodal input.

What license does the model use?

The model is released under the MIT license.

Other GGUFs of MiMo-V2.6-Pro-RL

Similar models

Browse all →

More from ggml-org

All by ggml-org →

More Mimo models

All Mimo →

Other large models

All large models →