Liquid AI · First MLX build 22 Jun 2026
LFM2.5-Embedding-350M
MLX
Text lfm1.0
A 354M parameter sentence similarity model from Liquid AI, with 3 MLX builds, from 4-bit to bf16.
- Parameters
- 354M From Hugging Face
- Quantisation
- 4-bit LFM2.5-Embedding-350M-4bit
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 199 MB Weights, estimated
- Downloads
- 720 This build, last 30 days
- Task
- Sentence similarity Text
- MLX builds
- 3 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 3 GB at 4-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 · Hugging Face CLI$ pip install -U huggingface_hub
$ hf download mlx-community/LFM2.5-Embedding-350M-4bitMLX builds
3 on mlx-community| LFM2.5-Embedding-350M-4bit | 4-bit | 199 MB | 3 GB | 720 |
|---|---|---|---|---|
| LFM2.5-Embedding-350M-8bit | 8-bit | 377 MB | 3 GB | 27 |
| LFM2.5-Embedding-350M-bf16 | bf16 | 709 MB | 3 GB | 335 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.