Model Intelligence Sheet
ggml-org/MiMo-V2.6-Flash-RL-GGUF overview
MiMo V2.6 Flash RL Run with https://llama.app bash llama serve hf ggml org/MiMo V2.6 Flash RL GGUF Source models https://huggingface.co/XiaomiMiMo/MiMo V2.6 Fl…
Runs locally from ~5.7 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
11 GGUF files detected
Direct downloads for local inference
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| MiMo-V2.6-Flash-RL-MXFP4-00001-of-00002.gguf | GGUF | GGUF | 5.7 MB | Download |
| MiMo-V2.6-Flash-RL-MXFP4-00002-of-00002.gguf | GGUF | GGUF | 155.87 GB | Download |
| MiMo-V2.6-Flash-RL-Q2_K-00001-of-00002.gguf | GGUF | Q2_K | 5.7 MB | Download |
| MiMo-V2.6-Flash-RL-Q2_K-00002-of-00002.gguf | GGUF | Q2_K | 117.54 GB | Download |
| dflash-MiMo-V2.6-Flash-RL-BF16.gguf | GGUF | BF16 | 2.74 GB | Download |
| dflash-MiMo-V2.6-Flash-RL-Q8_0.gguf | GGUF | Q8_0 | 1.46 GB | Download |
| mmproj-MiMo-V2.6-Flash-RL-BF16.gguf | GGUF | BF16 | 2.56 GB | Download |
| mmproj-MiMo-V2.6-Flash-RL-Q8_0.gguf | GGUF | Q8_0 | 1.46 GB | Download |
| mtp-MiMo-V2.6-Flash-RL-BF16.gguf | GGUF | BF16 | 4.17 GB | Download |
| mtp-MiMo-V2.6-Flash-RL-Q4_0.gguf | GGUF | Q4_0 | 1.18 GB | Download |
| mtp-MiMo-V2.6-Flash-RL-Q8_0.gguf | GGUF | Q8_0 | 2.22 GB | Download |
Model Details
| Model ID | ggml-org/MiMo-V2.6-Flash-RL-GGUF |
|---|---|
| Author | ggml-org |
| Pipeline | image-text-to-text |
| License | mit |
| Base model | XiaomiMiMo/MiMo-V2.6-Flash-RL |
| Last modified | 2026-09-25T13:55:37.000Z |
Model README
---
license: mit
pipeline_tag: image-text-to-text
tags:
- gguf
- quantized
base_model:
- XiaomiMiMo/MiMo-V2.6-Flash-RL
---
MiMo-V2.6-Flash-RL
Run with https://llama.app
llama serve -hf ggml-org/MiMo-V2.6-Flash-RL-GGUFSource models
- https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL
Notes
- The MXFP4 output keeps the routed experts at their native MXFP4 precision.
- The Q2_K output keeps the expert down projections at MXFP4, and quantizes the gate/up projections to Q2_K.
- Includes MTP sidecars (Q4_0 and Q8_0) for speculative decoding (
--mtp). - Includes a DFlash drafter sidecar (BF16 and Q8_0) for speculative decoding, converted from the
dflash/subdirectory of the source repo. - Includes a Q8_0 mmproj for the vision and audio encoders.
- Currently, the Q2 models do not use an imatrix calibration due to lack of one.
> [!IMPORTANT]
> This model is automatically converted using https://github.com/ggml-org/convert
Run ggml-org/MiMo-V2.6-Flash-RL-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models