Liquid AI · First MLX build 20 Jan 2026
LFM2.5-1.2B-Thinking
MLX
Text lfm1.0
A 1.2B parameter text generation model from Liquid AI, with 6 MLX builds, from 4-bit to bf16.
- Parameters
- 1.2B From Hugging Face
- Quantisation
- 6-bit LFM2.5-1.2B-Thinking-6bit
- Memory needed
- 4 GB Estimate, unified memory
- Download
- 951 MB Weights, estimated
- Downloads
- 686 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 6 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 4 GB at 6-bit
8 GB
16 GB
24 GB
32 GB
64 GB
128 GB
Fits with headroomTight, close other appsWon’t fit
Estimated from the size of the weights, plus a fifth for context and working memory and 2 GB for macOS and your apps. Long contexts need more.
Run it
Terminal · mlx-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/LFM2.5-1.2B-Thinking-6bit --prompt "Hello"MLX builds
6 on mlx-community| LFM2.5-1.2B-Thinking-OptiQ-4bit | 4-bit | 659 MB | 3 GB | 292 |
|---|---|---|---|---|
| LFM2.5-1.2B-Thinking-4bit | 4-bit | 659 MB | 3 GB | 196 |
| LFM2.5-1.2B-Thinking-5bit | 5-bit | 805 MB | 3 GB | 31 |
| LFM2.5-1.2B-Thinking-6bit | 6-bit | 951 MB | 4 GB | 686 |
| LFM2.5-1.2B-Thinking-8bit | 8-bit | 1.2 GB | 4 GB | 46 |
| LFM2.5-1.2B-Thinking-bf16 | bf16 | 2.3 GB | 5 GB | 39 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.