Hugging FaceFor programmers who want help reading logs, tracking down bugs and refactoring code.
- Rating
- 4.1
- Downloads
- 15K
- Parameters
- 229B, 10B active
- Context
- 192K
- Download
- 61 to 457 GB
- Commercial use
- No
Too big for this Mac
Its smallest size needs 120 GB of memory, more than M4 · 16 GB can give it.
About
Built to work alongside other agents, each one keeping a steady role and making its own decisions. Made to trace a server problem to its cause by checking monitoring figures, traces and databases.
How it compares
Of the 6 chat models Outmake recommends for a 128 GB Mac, it is the only one that can't read pictures.
Of the chat models Outmake recommends for a 128 GB Mac, it ranks 4th of 6 for speed, and Qwen3.6 35B is about 3.3 times as fast.
In LM Arena's public vote it rates 4th of the 4 chat models Outmake recommends for a 128 GB Mac that LM Arena rates.
Good to know
An internal version reworked a programming tool on its own over more than 100 rounds, choosing which changes to keep.
During development it updated its own memory and changed how it learned.
Sources: MiniMaxAI/MiniMax-M2.7 model card, Outmake's speed estimates, LM Arena and Hugging Face. Every line comes from one of these.
Ratings
4.2 out of 5, from how people voted on its answers.
LM Arena ratings of October 28, 2026 (CC BY 4.0): people vote between two answers without knowing which model wrote which. The bar ends at 1488, the best any open model has. The Arena tests the full model; a small size on a Mac answers a little worse.
Good at
Licence
MiniMax M2.7 Non-Commercial License. Commercial use: no. The maker's page has the full terms.
MiniMax M2.7 runs well from a Mac with an M2 Max and 96 GB of memory, at UD-IQ2_XXS, writing about 12 words (16 tokens) a second. 15 of 59 Mac configurations run it well.
Compare with
The nearest models in size, judged for the same Mac.
- DeepSeek V4 FlashLong documents and tool useToo big for this Mac · 128 GBGet
- Qwen3.8 Flash NextPreview of Qwen's next designToo big for this Mac · 91 GBGet
- Qwen3.5 397BFor conversations that mix text and pictures, with room to read a long report in one go.Too big for this Mac · 124 GBGet
- Qwen3.5 122BFor long documents and questions in many languages, and it can read photos and screenshots too.Too big for this Mac · 58 GBGet
Sizes
| Size | Download | Memory at 8K | Memory at 32K | Open |
|---|---|---|---|---|
| UD-Q3_K_XLSmaller | 102 GB | 103 GB | 106 GB | Open |
| UD-Q4_K_XLBalanced | 141 GB | 142 GB | 145 GB | Open |
| UD-Q5_K_XLBest | 169 GB | 170 GB | 174 GB | Open |
All 23 sizes
| Size | Download | Memory at 8K | Memory at 32K | Open |
|---|---|---|---|---|
| UD-IQ1_M | 61 GB | 62 GBestimated | 65 GB | Open |
| UD-IQ2_XXS | 65 GB | 66 GBestimated | 70 GB | Open |
| UD-IQ2_M | 70 GB | 71 GBestimated | 74 GB | Open |
| UD-Q2_K_XL | 75 GB | 76 GBestimated | 79 GB | Open |
| UD-IQ3_XXS | 80 GB | 81 GBestimated | 84 GB | Open |
| UD-IQ3_S | 84 GB | 84 GBestimated | 88 GB | Open |
| UD-Q3_K_S | 94 GB | 94 GBestimated | 98 GB | Open |
| UD-Q3_K_M | 101 GB | 102 GBestimated | 105 GB | Open |
| UD-IQ4_XS | 108 GB | 109 GBestimated | 113 GB | Open |
| UD-IQ4_NL | 111 GB | 112 GBestimated | 115 GB | Open |
| UD-Q4_K_S | 131 GB | 132 GBestimated | 135 GB | Open |
| MXFP4_MOE | 136 GB | 137 GBestimated | 140 GB | Open |
| UD-Q4_K_M | 140 GB | 141 GBestimated | 144 GB | Open |
| UD-Q5_K_S | 159 GB | 160 GBestimated | 163 GB | Open |
| UD-Q5_K_M | 169 GB | 170 GBestimated | 173 GB | Open |
| UD-Q6_K | 188 GB | 189 GBestimated | 192 GB | Open |
| UD-Q6_K_XL | 207 GB | 208 GBestimated | 212 GB | Open |
| Q8_0 | 243 GB | 244 GBestimated | 247 GB | Open |
| UD-Q8_K_XL | 247 GB | 247 GBestimated | 251 GB | Open |
| BF16 | 457 GB | 458 GBestimated | 462 GB | Open |
Memory is what llama.cpp's own calculator reported for each size when the catalog was built, with the app's settings. Smaller, Balanced and Best are the three sizes the store names.
Run It Anywhere
MiniMax M2.7 is an open model, so any app that runs GGUF files can run it. These take the UD-Q4_K_XL size from unsloth straight from Hugging Face.
- LM StudioMac, Windows, Linux
Open - JanMac, Windows, Linux
Open - MstyMac, Windows, Linux
Open - OllamaMac, Windows, Linux
ollama run hf.co/unsloth/MiniMax-M2.7-GGUF:UD-Q4_K_XL - llama.cppMac, Windows, Linux
llama-server -hf unsloth/MiniMax-M2.7-GGUF:UD-Q4_K_XL - Docker Model RunnerMac, Windows, Linux
docker model run hf.co/unsloth/MiniMax-M2.7-GGUF:UD-Q4_K_XL
On a PC Graphics Card
| Video memory | Largest size that fits | Download | Memory at 8K |
|---|---|---|---|
| 6 GB | None worth running | needs 71 GB | |
| 8 GB | None worth running | needs 71 GB | |
| 10 GB | None worth running | needs 71 GB | |
| 11 GB | None worth running | needs 71 GB | |
| 12 GB | None worth running | needs 71 GB | |
| 16 GB | None worth running | needs 71 GB | |
| 20 GB | None worth running | needs 71 GB | |
| 24 GB | None worth running | needs 71 GB | |
| 32 GB | None worth running | needs 71 GB | |
| 48 GB | None worth running | needs 71 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.
Tests
MMLU-Pro: graduate-level questions in 14 subjects. GPQA Diamond: questions PhD scientists write for each other. Humanity's Last Exam: questions written to be beyond today's models. Terminal-Bench 2.0: jobs finished alone in a terminal. AIME 2026: this year's AIME competition problems. Scores are the ones filed on the model's Hugging Face page or the board's own table, read 2026-11-05; some come from the maker.
On every Mac
| Mac | Size | Fit | Memory | Writes | Reads |
|---|---|---|---|---|---|
| M6 · 16 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M6 · 24 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M6 · 32 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M5 · 16 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M5 · 24 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M5 · 32 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M5 Pro · 24 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M5 Pro · 48 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M5 Pro · 64 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M5 Max · 36 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M5 Max · 48 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M5 Max · 64 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M5 Max · 128 GB | UD-Q3_K_S | Runs well | 98 GB | 19 | 813 |
| M5 Ultra · 96 GB | UD-IQ2_XXS | Runs well | 66 GB | 38 | 1300 |
| M5 Ultra · 256 GB | UD-Q5_K_XL | Runs great | 174 GB | 28 | 1625 |
| M5 Ultra · 512 GB | UD-Q5_K_XL | Runs great | 187 GB | 28 | 1625 |
| M4 · 16 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M4 · 24 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M4 · 32 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M4 Pro · 24 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M4 Pro · 48 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M4 Pro · 64 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M4 Max · 36 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M4 Max · 48 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M4 Max · 64 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M4 Max · 96 GB | UD-IQ2_XXS | Runs well | 66 GB | 12 | 218 |
| M4 Max · 128 GB | UD-Q3_K_S | Runs well | 98 GB | 14 | 273 |
| M3 · 8 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M3 · 16 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M3 · 24 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M3 Pro · 18 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M3 Pro · 36 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M3 Max · 36 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M3 Max · 48 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M3 Max · 64 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M3 Max · 96 GB | UD-IQ2_XXS | Runs well | 66 GB | 9.8 | 170 |
| M3 Max · 128 GB | UD-Q3_K_S | Runs well | 98 GB | 11 | 226 |
| M3 Ultra · 96 GB | UD-IQ2_XXS | Runs well | 66 GB | 20 | 339 |
| M3 Ultra · 256 GB | UD-Q5_K_XL | Runs great | 174 GB | 16 | 339 |
| M3 Ultra · 512 GB | UD-Q5_K_XL | Runs great | 187 GB | 16 | 452 |
| M2 · 8 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M2 · 16 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M2 · 24 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M2 Pro · 16 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M2 Pro · 32 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M2 Max · 32 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M2 Max · 64 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M2 Max · 96 GB | UD-IQ2_XXS | Runs well | 66 GB | 12 | 211 |
| M2 Ultra · 64 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M2 Ultra · 128 GB | UD-Q3_K_S | Runs well | 98 GB | 19 | 333 |
| M2 Ultra · 192 GB | UD-Q3_K_XL | Runs great | 120 GB | 18 | 333 |
| M1 · 8 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M1 · 16 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M1 Pro · 16 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M1 Pro · 32 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M1 Max · 32 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M1 Max · 64 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M1 Ultra · 64 GB | UD-Q3_K_XL | Too big | 120 GB | ||
| M1 Ultra · 128 GB | UD-Q3_K_S | Runs well | 98 GB | 15 | 247 |
Writes and reads are words a second (a token is about three quarters of a word), 4K tokens into a chat, estimated unless marked measured. The size is the one the store offers that Mac. Constants fitted October 11, 2026; how, and how wrong.
Badge for the model card
Shows on MiniMax M2.7's Hugging Face card which Macs run it well, linked to this page. It updates every morning.
Sizes and files are read from unsloth/MiniMax-M2.7-GGUF on Hugging Face and checked every morning. The rating is the model's score on the LM Arena text leaderboard (CC BY 4.0), where people vote between two answers without knowing which model wrote which. Catalog of November 4, 2026.