MiMo V2.6 Flash RL GSQ RCO needs about 132.3 GB of memory at its main file (a 115.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 2 quants, from 115.7 GB to 135.0 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 64.4 GB.
MiMo V2.6 Flash RL GSQ RCO – 309B text-only GGUF model
MiMo V2.6 Flash RL GSQ RCO is a 309‑billion‑parameter text model from the MiMo family. It is released as 3‑bit and 3.5‑bit GGUFs, with 115.68 GB and 134.98 GB file sizes respectively. The model supports text chat and thinking mode but excludes vision, audio and multi‑token prediction weights.
Good at
- Large context up to 1,048,576 tokens
- Text‑only chat with thinking mode
- Mixture‑of‑experts architecture
Limits
- Requires >130 GB VRAM for 3‑bit, >150 GB for 3.5‑bit
- No vision or audio support
- MIT license only
Best for
- High‑capacity text chat agents
- Long‑horizon reasoning tasks
- Research on RL‑based self‑improvement
Who it’s for. Researchers and developers needing a massive text‑only model with reinforcement‑learning‑derived capabilities. The model is suitable for environments that can accommodate its large memory footprint.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 115.7 GB | 132.3 GB | Too big | Get file → | |
| — | 135.0 GB | 151.6 GB | Too big | Get file → |
For 8‑k context, the 3‑bit GGUF uses about 132.3 GB of VRAM, while the 3.5‑bit uses about 151.6 GB. If your GPU has 128 GB, the 3‑bit model is the only feasible option. For 200 GB VRAM, either quantization can be used.
What GPU do you need for MiMo V2.6 Flash RL GSQ RCO?
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.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about MiMo V2.6 Flash RL GSQ RCO
How much VRAM does MiMo V2.6 Flash RL GSQ RCO need?
About 132.3 GB at its main file with an 8K context window. That is the 115.7 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 Flash RL GSQ RCO on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 132.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of MiMo V2.6 Flash RL GSQ RCO GGUF?
It depends on the quant: file is 115.7 GB, file is 135.0 GB.
What context length does MiMo V2.6 Flash RL GSQ RCO 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 16.1 GB at 8K and 64.4 GB at 32K.
Where can I download MiMo V2.6 Flash RL GSQ RCO GGUF?
The files are on Hugging Face in the pfeifferj/MiMo-V2.6-Flash-RL-GSQ-RCO-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 Flash RL GSQ RCO 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 Flash RL on 8 GB VRAM?
No. The model requires over 130 GB VRAM even with the smallest quantization.
Does this model support vision or audio?
No. The GGUFs exclude vision, audio and multi‑token prediction weights.
What license does the model use?
The model is released under the MIT license.