Outmake Models
Models

LFM2.5 1.2B

Liquid AI · January 2026

Get Hugging Face

A 1.2B chat model from Liquid AI.

Downloads
340K
Likes
291
Parameters
1.2B
Context
125K
Licence
other

Runs great on M4 · 16 GB

The Q8_0 size from LiquidAI is the one to get: a 1.2 GB download that uses 1.8 GB of memory, and writes about 42 words a second.

LFM2.5 1.2B runs well from a Mac with an M1 and 8 GB of memory. 59 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From LiquidAI/LFM2.5-1.2B-Instruct-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
Q4_00.7 GB0.9 GBestimatedOpen
Q4_K_M0.7 GB0.9 GBestimatedOpen
Q5_K_M0.8 GB1.0 GBestimatedOpen
Q6_K1.0 GB1.1 GBestimatedOpen
Q8_01.2 GB1.4 GBestimatedOpen
BF162.3 GB2.5 GBestimatedOpen
F162.3 GB2.5 GBestimatedOpen

Run It Anywhere

LFM2.5 1.2B is an open model, so any app that runs GGUF files can run it. These take the Q8_0 size from LiquidAI straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/LiquidAI/LFM2.5-1.2B-Instruct-GGUF:Q8_0
  • llama.cppMac, Windows, Linuxllama-server -hf LiquidAI/LFM2.5-1.2B-Instruct-GGUF:Q8_0
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/LiquidAI/LFM2.5-1.2B-Instruct-GGUF:Q8_0
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBF162.3 GB2.5 GB
8 GBF162.3 GB2.5 GB
10 GBF162.3 GB2.5 GB
11 GBF162.3 GB2.5 GB
12 GBF162.3 GB2.5 GB
16 GBF162.3 GB2.5 GB
20 GBF162.3 GB2.5 GB
24 GBF162.3 GB2.5 GB
32 GBF162.3 GB2.5 GB
48 GBF162.3 GB2.5 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 GBQ8_0Runs great1.8 GB65
M6 · 24 GBQ8_0Runs great1.8 GB65
M6 · 32 GBQ8_0Runs great1.8 GB65
M5 · 16 GBQ8_0Runs great1.8 GB60
M5 · 24 GBQ8_0Runs great1.8 GB60
M5 · 32 GBQ8_0Runs great1.8 GB60
M5 Pro · 24 GBQ8_0Runs great1.8 GB115
M5 Pro · 48 GBQ8_0Runs great1.8 GB115
M5 Pro · 64 GBQ8_0Runs great1.8 GB115
M5 Max · 36 GBQ8_0Runs great1.8 GB164
M5 Max · 48 GBQ8_0Runs great1.8 GB209
M5 Max · 64 GBQ8_0Runs great1.8 GB209
M5 Max · 128 GBQ8_0Runs great1.8 GB209
M5 Ultra · 96 GBQ8_0Runs great1.8 GB356
M5 Ultra · 256 GBQ8_0Runs great1.8 GB356
M5 Ultra · 512 GBQ8_0Runs great1.8 GB356
M4 · 16 GBQ8_0Runs great1.8 GB42
M4 · 24 GBQ8_0Runs great1.8 GB42
M4 · 32 GBQ8_0Runs great1.8 GB42
M4 Pro · 24 GBQ8_0Runs great1.8 GB82
M4 Pro · 48 GBQ8_0Runs great1.8 GB82
M4 Pro · 64 GBQ8_0Runs great1.8 GB82
M4 Max · 36 GBQ8_0Runs great1.8 GB108
M4 Max · 48 GBQ8_0Runs great1.8 GB129
M4 Max · 64 GBQ8_0Runs great1.8 GB129
M4 Max · 96 GBQ8_0Runs great1.8 GB108
M4 Max · 128 GBQ8_0Runs great1.8 GB129
M3 · 8 GBQ8_0Runs great1.8 GB37
M3 · 16 GBQ8_0Runs great1.8 GB37
M3 · 24 GBQ8_0Runs great1.8 GB37
M3 Pro · 18 GBQ8_0Runs great1.8 GB53
M3 Pro · 36 GBQ8_0Runs great1.8 GB53
M3 Max · 36 GBQ8_0Runs great1.8 GB92
M3 Max · 48 GBQ8_0Runs great1.8 GB113
M3 Max · 64 GBQ8_0Runs great1.8 GB113
M3 Max · 96 GBQ8_0Runs great1.8 GB92
M3 Max · 128 GBQ8_0Runs great1.8 GB113
M3 Ultra · 96 GBQ8_0Runs great1.8 GB171
M3 Ultra · 256 GBQ8_0Runs great1.8 GB171
M3 Ultra · 512 GBQ8_0Runs great1.8 GB171
M2 · 8 GBQ8_0Runs great1.8 GB37
M2 · 16 GBQ8_0Runs great1.8 GB37
M2 · 24 GBQ8_0Runs great1.8 GB37
M2 Pro · 16 GBQ8_0Runs great1.8 GB67
M2 Pro · 32 GBQ8_0Runs great1.8 GB67
M2 Max · 32 GBQ8_0Runs great1.8 GB112
M2 Max · 64 GBQ8_0Runs great1.8 GB112
M2 Max · 96 GBQ8_0Runs great1.8 GB112
M2 Ultra · 64 GBQ8_0Runs great1.8 GB169
M2 Ultra · 128 GBQ8_0Runs great1.8 GB169
M2 Ultra · 192 GBQ8_0Runs great1.8 GB169
M1 · 8 GBQ8_0Runs great1.8 GB25
M1 · 16 GBQ8_0Runs great1.8 GB25
M1 Pro · 16 GBQ8_0Runs great1.8 GB59
M1 Pro · 32 GBQ8_0Runs great1.8 GB59
M1 Max · 32 GBQ8_0Runs great1.8 GB93
M1 Max · 64 GBQ8_0Runs great1.8 GB93
M1 Ultra · 64 GBQ8_0Runs great1.8 GB129
M1 Ultra · 128 GBQ8_0Runs great1.8 GB129

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

Packaged by

4 repositories on Hugging Face carry this model for llama.cpp. Outmake reads them all as one model.

Information

Maker
Liquid AI
Released
January 2026
Parameters
1.2B
Longest chat
125K tokens
Reads pictures
No
Licence
other
Downloads this month
339,899
Likes
291
Sizes
7, 0.7 to 2.3 GB
Kind
Maker's release

Badge for the model card

Shows on LFM2.5 1.2B'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