Hugging FaceA 310B, 6.2B active chat model from Xiaomi that reads pictures as well as text.
- Downloads
- 47K
- Likes
- 49
- Parameters
- 310B, 6.2B active
- Context
- 1M
- Licence
- mit
Too big for this Mac
Its smallest size needs 122 GB of memory, more than M4 · 16 GB can give it.
MiMo V2.6 Flash RL runs well from a Mac with an M2 Ultra and 192 GB of memory. 5 of 59 Mac configurations run it well.
Compare with
The nearest models in size.
- MiMo V2.5Xiaomi · 3.6K downloads · Arena 1434Too big for this Mac · 93 GB
- MiMo V2 FlashXiaomi · 21K downloads · Arena 1386Too big for this Mac · 78 GB
- Hy3Tencent · 147K downloads · Arena 1454Too big for this Mac · 64 GB
- Llama 3 1 Nemotron Ultra 253B V1NVIDIA · 2.2K downloads · Arena 1348Too big for this Mac · 54 GB
Sizes
From ggml-org/MiMo-V2.6-Flash-RL-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.
Run It Anywhere
MiMo V2.6 Flash RL is an open model, so any app that runs GGUF files can run it. These take the Q2_K size from ggml-org straight from Hugging Face.
- LM StudioMac, Windows, Linux
Open - JanMac, Windows, Linux
Open - MstyMac, Windows, Linux
Open - OllamaMac, Windows, Linux
ollama run hf.co/ggml-org/MiMo-V2.6-Flash-RL-GGUF:Q2_K - llama.cppMac, Windows, Linux
llama-server -hf ggml-org/MiMo-V2.6-Flash-RL-GGUF:Q2_K - Docker Model RunnerMac, Windows, Linux
docker model run hf.co/ggml-org/MiMo-V2.6-Flash-RL-GGUF:Q2_K
On a PC Graphics Card
| Video memory | Largest size that fits | Download | Memory at 8K |
|---|---|---|---|
| 6 GB | None worth running | needs 168 GB | |
| 8 GB | None worth running | needs 168 GB | |
| 10 GB | None worth running | needs 168 GB | |
| 11 GB | None worth running | needs 168 GB | |
| 12 GB | None worth running | needs 168 GB | |
| 16 GB | None worth running | needs 168 GB | |
| 20 GB | None worth running | needs 168 GB | |
| 24 GB | None worth running | needs 168 GB | |
| 32 GB | None worth running | needs 168 GB | |
| 48 GB | None worth running | needs 168 GB |
The largest size, of those that keep three bits a weight or more, whose memory at an 8K chat is under 92% of the card's; the rest is the desktop's. Estimated from the files, the same way as for a Mac; speed on a PC is not estimated yet. A size that doesn't fit the card still runs, slower, with part of it in ordinary memory.
On every Mac
| Mac | Size | Fit | Memory | Writes |
|---|---|---|---|---|
| M6 · 16 GB | IQ2_M | Too big | 122 GB | |
| M6 · 24 GB | IQ2_M | Too big | 122 GB | |
| M6 · 32 GB | IQ2_M | Too big | 122 GB | |
| M5 · 16 GB | IQ2_M | Too big | 122 GB | |
| M5 · 24 GB | IQ2_M | Too big | 122 GB | |
| M5 · 32 GB | IQ2_M | Too big | 122 GB | |
| M5 Pro · 24 GB | IQ2_M | Too big | 122 GB | |
| M5 Pro · 48 GB | IQ2_M | Too big | 122 GB | |
| M5 Pro · 64 GB | IQ2_M | Too big | 122 GB | |
| M5 Max · 36 GB | IQ2_M | Too big | 122 GB | |
| M5 Max · 48 GB | IQ2_M | Too big | 122 GB | |
| M5 Max · 64 GB | IQ2_M | Too big | 122 GB | |
| M5 Max · 128 GB | IQ2_M | Too big | 122 GB | |
| M5 Ultra · 96 GB | IQ2_M | Too big | 122 GB | |
| M5 Ultra · 256 GB | MXFP4 | Runs great | 174 GB | 89 |
| M5 Ultra · 512 GB | MXFP4 | Runs great | 181 GB | 89 |
| M4 · 16 GB | IQ2_M | Too big | 122 GB | |
| M4 · 24 GB | IQ2_M | Too big | 122 GB | |
| M4 · 32 GB | IQ2_M | Too big | 122 GB | |
| M4 Pro · 24 GB | IQ2_M | Too big | 122 GB | |
| M4 Pro · 48 GB | IQ2_M | Too big | 122 GB | |
| M4 Pro · 64 GB | IQ2_M | Too big | 122 GB | |
| M4 Max · 36 GB | IQ2_M | Too big | 122 GB | |
| M4 Max · 48 GB | IQ2_M | Too big | 122 GB | |
| M4 Max · 64 GB | IQ2_M | Too big | 122 GB | |
| M4 Max · 96 GB | IQ2_M | Too big | 122 GB | |
| M4 Max · 128 GB | IQ2_M | Too big | 122 GB | |
| M3 · 8 GB | IQ2_M | Too big | 122 GB | |
| M3 · 16 GB | IQ2_M | Too big | 122 GB | |
| M3 · 24 GB | IQ2_M | Too big | 122 GB | |
| M3 Pro · 18 GB | IQ2_M | Too big | 122 GB | |
| M3 Pro · 36 GB | IQ2_M | Too big | 122 GB | |
| M3 Max · 36 GB | IQ2_M | Too big | 122 GB | |
| M3 Max · 48 GB | IQ2_M | Too big | 122 GB | |
| M3 Max · 64 GB | IQ2_M | Too big | 122 GB | |
| M3 Max · 96 GB | IQ2_M | Too big | 122 GB | |
| M3 Max · 128 GB | IQ2_M | Too big | 122 GB | |
| M3 Ultra · 96 GB | IQ2_M | Too big | 122 GB | |
| M3 Ultra · 256 GB | MXFP4 | Runs great | 174 GB | 42 |
| M3 Ultra · 512 GB | MXFP4 | Runs great | 181 GB | 42 |
| M2 · 8 GB | IQ2_M | Too big | 122 GB | |
| M2 · 16 GB | IQ2_M | Too big | 122 GB | |
| M2 · 24 GB | IQ2_M | Too big | 122 GB | |
| M2 Pro · 16 GB | IQ2_M | Too big | 122 GB | |
| M2 Pro · 32 GB | IQ2_M | Too big | 122 GB | |
| M2 Max · 32 GB | IQ2_M | Too big | 122 GB | |
| M2 Max · 64 GB | IQ2_M | Too big | 122 GB | |
| M2 Max · 96 GB | IQ2_M | Too big | 122 GB | |
| M2 Ultra · 64 GB | IQ2_M | Too big | 122 GB | |
| M2 Ultra · 128 GB | IQ2_M | Too big | 122 GB | |
| M2 Ultra · 192 GB | Q2_K | Runs great | 130 GB | 41 |
| M1 · 8 GB | IQ2_M | Too big | 122 GB | |
| M1 · 16 GB | IQ2_M | Too big | 122 GB | |
| M1 Pro · 16 GB | IQ2_M | Too big | 122 GB | |
| M1 Pro · 32 GB | IQ2_M | Too big | 122 GB | |
| M1 Max · 32 GB | IQ2_M | Too big | 122 GB | |
| M1 Max · 64 GB | IQ2_M | Too big | 122 GB | |
| M1 Ultra · 64 GB | IQ2_M | Too big | 122 GB | |
| M1 Ultra · 128 GB | IQ2_M | Too big | 122 GB |
Writes is words a second, estimated from the Mac's memory speed and the model's size; how, and how wrong.
Packaged by
5 repositories on Hugging Face carry this model for llama.cpp. Outmake reads them all as one model.
Information
- Maker
- Xiaomi
- Released
- September 2026
- Parameters
- 310B, 6.2B active
- Longest chat
- 1M tokens
- Reads pictures
- Yes
- Licence
- mit
- Downloads this month
- 46,851
- Likes
- 49
- Sizes
- 2, 126 to 167 GB
- Kind
- Maker's release
Badge for the model card
Shows on MiMo V2.6 Flash RL's Hugging Face card which Macs run it well, linked to this page. It updates every morning.
Files, downloads and likes read from Hugging Face on November 5, 2026 and refreshed every morning. Memory and speed are estimates until someone running Outmake measures this model on that Mac.