Outmake Models
Models

Gemmable 4 12B MTP

Mia-AiLab · June 2026

Get Hugging Face

A 12B chat model from Mia-AiLab.

Downloads
238K
Likes
53
Parameters
12B
Context
256K

Runs well on M4 · 16 GB

The Q6_K size from Mia-AiLab is the one to get: a 9.8 GB download that uses 10 GB of memory, and writes about 6.4 words a second.

Gemmable 4 12B MTP runs well from a Mac with an M1 and 16 GB of memory. 56 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From Mia-AiLab/Gemmable-4-12B-MTP-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
Q4_K_M7.4 GB7.7 GBestimatedOpen
Q6_K9.8 GB10 GBestimatedOpen
Q8_013 GB13 GBestimatedOpen

Run It Anywhere

Gemmable 4 12B MTP is an open model, so any app that runs GGUF files can run it. These take the Q6_K size from Mia-AiLab straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/Mia-AiLab/Gemmable-4-12B-MTP-GGUF:Q6_K
  • llama.cppMac, Windows, Linuxllama-server -hf Mia-AiLab/Gemmable-4-12B-MTP-GGUF:Q6_K
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/Mia-AiLab/Gemmable-4-12B-MTP-GGUF:Q6_K
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBNone worth runningneeds 7.7 GB
8 GBQ4_K_M7.4 GB7.7 GB
10 GBQ4_K_M7.4 GB7.7 GB
11 GBQ6_K9.8 GB10 GB
12 GBQ6_K9.8 GB10 GB
16 GBQ8_013 GB13 GB
20 GBQ8_013 GB13 GB
24 GBQ8_013 GB13 GB
32 GBQ8_013 GB13 GB
48 GBQ8_013 GB13 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
MacSizeFitMemoryWrites
M6 · 16 GBQ6_KRuns well10 GB9.5
M6 · 24 GBQ6_KRuns great11 GB9.5
M6 · 32 GBQ8_0Runs great14 GB8.5
M5 · 16 GBQ6_KRuns well10 GB8.6
M5 · 24 GBQ6_KRuns great11 GB8.6
M5 · 32 GBQ8_0Runs great14 GB7.7
M5 Pro · 24 GBQ6_KRuns great11 GB17
M5 Pro · 48 GBQ8_0Runs great16 GB15
M5 Pro · 64 GBQ8_0Runs great16 GB15
M5 Max · 36 GBQ8_0Runs great16 GB22
M5 Max · 48 GBQ8_0Runs great16 GB29
M5 Max · 64 GBQ8_0Runs great16 GB29
M5 Max · 128 GBQ8_0Runs great16 GB29
M5 Ultra · 96 GBQ8_0Runs great16 GB54
M5 Ultra · 256 GBQ8_0Runs great16 GB54
M5 Ultra · 512 GBQ8_0Runs great16 GB54
M4 · 16 GBQ6_KRuns well10 GB6.4
M4 · 24 GBQ6_KRuns great11 GB6.4
M4 · 32 GBQ8_0Runs great14 GB5.8
M4 Pro · 24 GBQ6_KRuns great11 GB13
M4 Pro · 48 GBQ8_0Runs great16 GB12
M4 Pro · 64 GBQ8_0Runs great16 GB12
M4 Max · 36 GBQ8_0Runs great16 GB17
M4 Max · 48 GBQ8_0Runs great16 GB22
M4 Max · 64 GBQ8_0Runs great16 GB22
M4 Max · 96 GBQ8_0Runs great16 GB17
M4 Max · 128 GBQ8_0Runs great16 GB22
M3 · 8 GBQ4_K_MToo big10 GB
M3 · 16 GBQ6_KRuns well10 GB5.4
M3 · 24 GBQ6_KRuns great11 GB5.4
M3 Pro · 18 GBQ4_K_MRuns great8.3 GB10
M3 Pro · 36 GBQ8_0Runs great16 GB7.2
M3 Max · 36 GBQ8_0Runs great16 GB14
M3 Max · 48 GBQ8_0Runs great16 GB17
M3 Max · 64 GBQ8_0Runs great16 GB17
M3 Max · 96 GBQ8_0Runs great16 GB14
M3 Max · 128 GBQ8_0Runs great16 GB17
M3 Ultra · 96 GBQ8_0Runs great16 GB30
M3 Ultra · 256 GBQ8_0Runs great16 GB30
M3 Ultra · 512 GBQ8_0Runs great16 GB30
M2 · 8 GBQ4_K_MToo big10 GB
M2 · 16 GBQ6_KRuns well10 GB5.4
M2 · 24 GBQ6_KRuns great11 GB5.4
M2 Pro · 16 GBQ6_KRuns well10 GB10
M2 Pro · 32 GBQ8_0Runs great14 GB9.4
M2 Max · 32 GBQ8_0Runs great14 GB17
M2 Max · 64 GBQ8_0Runs great16 GB17
M2 Max · 96 GBQ8_0Runs great16 GB17
M2 Ultra · 64 GBQ8_0Runs great16 GB30
M2 Ultra · 128 GBQ8_0Runs great16 GB30
M2 Ultra · 192 GBQ8_0Runs great16 GB30
M1 · 8 GBQ4_K_MToo big10 GB
M1 · 16 GBQ6_KRuns well10 GB3.7
M1 Pro · 16 GBQ6_KRuns well10 GB9.7
M1 Pro · 32 GBQ8_0Runs great14 GB8.9
M1 Max · 32 GBQ8_0Runs great14 GB16
M1 Max · 64 GBQ8_0Runs great16 GB16
M1 Ultra · 64 GBQ8_0Runs great16 GB26
M1 Ultra · 128 GBQ8_0Runs great16 GB26

Writes is words a second, estimated from the Mac's memory speed and the model's size; how, and how wrong.

Information

Maker
Mia-AiLab
Released
June 2026
Parameters
12B
Longest chat
256K tokens
Reads pictures
No
Licence
Not stated
Downloads this month
237,696
Likes
53
Sizes
3, 7.4 to 13 GB
Kind
Maker's release

Badge for the model card

Shows on Gemmable 4 12B MTP's Hugging Face card which Macs run it well, linked to this page. It updates every morning.

Runs on Mac badge

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.

Your Computer

Every model is judged for this Mac. Both are under About This Mac, in the Apple menu.

Every Mac · Every graphics card

Civitai

Open in OutmakeHugging Face

New to Outmake? Download the free beta