MiMo V2.6 Flash RL UNCENSORED IQ2S Down needs about 109.3 GB of memory at IQ2_XXS (a 91.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds. Longer context needs more: at 32K tokens the KV cache alone adds about 68.5 GB.
MiMo V2.6 Flash RL UNCENSORED IQ2_XXS – aggressively quantized local LLM
MiMo V2.6 Flash RL UNCENSORED is a 310B‑parameter mixture‑of‑experts model with 1 M context length. It includes vision, audio, video, and a DFlash speculative‑decoding head. The IQ2_XXS quantization is designed for local llama.cpp inference.
Good at
- Reasoning
- Tools usage
- Large context handling
Limits
- Aggressive quantization increases perplexity
- Requires 109.3 GB VRAM at 8 k context
- No compliance‑tuning internals shipped
Best for
- Local inference on high‑memory GPUs
- Rapid prototyping with reduced model size
- Testing quantization trade‑offs
Who it’s for. Developers and researchers who need a local MiMo model with vision, audio, and video capabilities but can tolerate higher perplexity for reduced memory usage. The model is MIT‑licensed and supports both thinking‑ON and thinking‑OFF modes.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ2_XXS | 91.7 GB (4 parts) | 109.3 GB | Too big | Get file → |
For 8 k context, the IQ2_XXS quantization requires approximately 109.3 GB VRAM. If your GPU has less memory, consider using a higher‑precision quant such as Q8_0 or Q5_K for the attention tensors, but note that the full model will still be large. The file size is 91.7 GB, split into four shards.
What GPU do you need for MiMo V2.6 Flash RL UNCENSORED IQ2S Down?
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 Flash RL UNCENSORED IQ2S Down quant should you choose?
IQ2_XXS (91.7 GB) is the middle option in this repository. 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
Source
Frequently asked questions about MiMo V2.6 Flash RL UNCENSORED IQ2S Down
How much VRAM does MiMo V2.6 Flash RL UNCENSORED IQ2S Down need?
About 109.3 GB at IQ2_XXS with an 8K context window. That is the 91.7 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 UNCENSORED IQ2S Down on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 109.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of MiMo V2.6 Flash RL UNCENSORED IQ2S Down GGUF?
It depends on the quant: IQ2_XXS is 91.7 GB.
What context length does MiMo V2.6 Flash RL UNCENSORED IQ2S Down 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 V2.6 Flash RL UNCENSORED IQ2S Down GGUF?
The files are on Hugging Face in the BattleGhost/MiMo-V2.6-Flash-RL-UNCENSORED-IQ2S-Down-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 UNCENSORED IQ2S Down 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 UNCENSORED on 8 GB VRAM?
No. The IQ2_XXS build requires about 109.3 GB VRAM at 8 k context, far above 8 GB.
Does this model include compliance tuning?
No. The repository contains only weights, config, tokenizer, and assets; no compliance‑tuning internals or refusal vectors are shipped.
What is the context length of this model?
The model supports up to 1 048 576 tokens of context.