LFM2 Models — Hardware Requirements
9 LFM2 models from Liquid AI and the community, from the smallest that runs in 0.8 GB of VRAM at Q4 up to 23.8B parameters. Every row links to full quantization tables and GPU compatibility.
All LFM2 Models by Size
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| LFM2 350M | 354M | 0.8 GB | 128K | ||
| LFM2 1.2B | 1.2B | 1.2 GB | 128K | ||
| LFM2 VL 1.6B | 1.6B | 1.5 GB | 128K | ||
| LFM2 2.6B Longevity | 2.6B | 2.3 GB | 128K | ||
| LFM2 8B A1B | 8.3B | 5.5 GB | 128K |
Community variants and fine-tunes (4)
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| LFM2 350M Extract | 354M | 0.8 GB | 128K | ||
| LFM2 1.2B RAG | 1.2B | 1.2 GB | 128K | ||
| LFM2 1.2B Extract | 1.2B | 1.2 GB | 128K | ||
| LFM2 24B A2B | 23.8B | 14.9 GB | 128K |
Frequently Asked Questions
- How much VRAM do I need to run a LFM2 model?
- The smallest LFM2 model, LFM2 350M, runs from 0.8 GB of VRAM at Q4_K_M (smaller low-bit quantizations start at 0.5 GB with a noticeable quality loss). Larger family members need proportionally more — see the table above for every model.
- Which LFM2 models can I run on a 16 GB GPU?
- 9 of 9 LFM2 models fit in 16 GB of VRAM at Q4_K_M, including LFM2 24B A2B, LFM2 1.2B, LFM2 VL 1.6B.
- What is the most popular LFM2 model to run locally?
- LFM2 24B A2B is the most downloaded LFM2 model in local-friendly quantized formats. It runs from 14.9 GB of VRAM at Q4_K_M.