8-bit Open on Hugging Face
Parameters
2.6B
Weights (est.)
≈ 2.7 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-lm on an Apple silicon Mac:
pip install mlx-lmmlx_lm.generate --model mlx-community/LFM2-2.6B-Exp-8bit --prompt "Hello"
Details
- Base model
- LiquidAI/LFM2-2.6B-Exp (quantized)
- Vendor
- LiquidAI
- Task
- Text generation
- Quantisation
- 8-bit
- Licence
- lfm1.0
- Downloads
- 31 in the last 30 days · 1.8K all time · 2 likes
- Published
- 2025-12-26 · updated 2025-12-26 (10mo ago)
- Languages
- en, ar, zh, fr, de, ja, ko, es
- Tags
- lfm2, liquid, edge
Other MLX builds of LiquidAI/LFM2-2.6B-Exp
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| LFM2-2.6B-Exp-4bit | 4-bit | ≈ 1.4 GB | 8 GB+ | 320 |
| LFM2-2.6B-Exp-5bit | 5-bit | ≈ 1.8 GB | 8 GB+ | 17 |
| LFM2-2.6B-Exp-6bit | 6-bit | ≈ 2.1 GB | 8 GB+ | 39 |
| LFM2-2.6B-Exp-8bit (this one) | 8-bit | ≈ 2.7 GB | 8 GB+ | 31 |
| LFM2-2.6B-Exp-bf16 | bf16 | ≈ 5.1 GB | 8 GB+ | 35 |