8-bit Open on Hugging Face
Parameters
1.6B
Weights (est.)
≈ 2.1 GB
Mac memory (est.)
8 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-vlm on an Apple silicon Mac:
pip install mlx-vlmpython -m mlx_vlm.generate --model mlx-community/LFM2.5-VL-1.6B-8bit --image photo.jpg --prompt "Describe this image"
Details
- Base model
- LiquidAI/LFM2.5-1.2B-Base (quantized)
- Vendor
- LiquidAI
- Task
- Vision + text
- Quantisation
- 8-bit
- Licence
- lfm1.0
- Downloads
- 110 in the last 30 days · 1.2K all time · 2 likes
- Published
- 2026-01-06 · updated 2026-01-06 (10mo ago)
- Languages
- en, ja, ko, fr, es, de, ar, zh
- Tags
- lfm2_vl, liquid, lfm2, lfm2-vl, edge, lfm2.5-vl, lfm2.5
Other MLX builds of LiquidAI/LFM2.5-1.2B-Base
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| LFM2.5-1.2B-Base-4bit | 4-bit | ≈ 659 MB | 8 GB+ | 30 |
| LFM2.5-1.2B-Base-5bit | 5-bit | ≈ 805 MB | 8 GB+ | 21 |
| LFM2.5-1.2B-Base-6bit | 6-bit | ≈ 951 MB | 8 GB+ | 20 |
| LFM2.5-1.2B-Base-8bit | 8-bit | ≈ 1.2 GB | 8 GB+ | 17 |
| LFM2.5-VL-1.6B-4bit | 4-bit | ≈ 1.5 GB | 8 GB+ | 3.1K |
| LFM2.5-VL-1.6B-5bit | 5-bit | ≈ 1.6 GB | 8 GB+ | 27 |
| LFM2.5-VL-1.6B-6bit | 6-bit | ≈ 1.8 GB | 8 GB+ | 28 |
| LFM2.5-VL-1.6B-8bit (this one) | 8-bit | ≈ 2.1 GB | 8 GB+ | 110 |
| LFM2.5-1.2B-Base-bf16 | bf16 | ≈ 2.3 GB | 8 GB+ | 23 |
| LFM2.5-VL-1.6B-bf16 | bf16 | ≈ 3.2 GB | 8 GB+ | 29 |