6-bit Open on Hugging Face
Parameters
24B
Weights (est.)
≈ 19.5 GB
Mac memory (est.)
24 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-lm on an Apple silicon Mac:
pip install mlx-lmmlx_lm.generate --model mlx-community/LFM2.5-24B-A2B-Preview-6bit --prompt "Hello"
Details
- Base model
- lq-preview/LFM2.5-24B-A2B-Preview (quantized)
- Vendor
- lq-preview
- Task
- Text generation
- Quantisation
- 6-bit
- Licence
- lfm1.0
- Downloads
- 270 in the last 30 days · 270 all time · 0 likes
- Published
- 2026-10-13 · updated 2026-10-13 (29d ago)
- Languages
- en, ar, zh, fr, de, ja, ko, es, pt, it
- Tags
- lfm2_moe, mlx-vlm, lfm2.5
Other MLX builds of lq-preview/LFM2.5-24B-A2B-Preview
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| LFM2.5-24B-A2B-Preview-mxfp4 | MXFP4 | ≈ 13.5 GB | 24 GB+ | 402 |
| LFM2.5-24B-A2B-Preview-4bit | 4-bit | ≈ 13.5 GB | 24 GB+ | 392 |
| LFM2.5-24B-A2B-Preview-nvfp4 | 4-bit | ≈ 13.5 GB | 24 GB+ | 277 |
| LFM2.5-24B-A2B-Preview-5bit | 5-bit | ≈ 16.5 GB | 24 GB+ | 292 |
| LFM2.5-24B-A2B-Preview-6bit (this one) | 6-bit | ≈ 19.5 GB | 24 GB+ | 270 |
| LFM2.5-24B-A2B-Preview-8bit | 8-bit | ≈ 25.5 GB | 32 GB+ | 191 |
| LFM2.5-24B-A2B-Preview-mxfp8 | MXFP8 | ≈ 25.5 GB | 32 GB+ | 155 |
| LFM2.5-24B-A2B-Preview-bf16 | bf16 | ≈ 47.9 GB | 64 GB+ | 722 |