LFM2 24B A2B — 24B MoE.
Run LFM2 24B A2B locally with Ollama — needs 15.2 GB VRAM at Q4_K_M:
ollama run lfm2:24b| Quant | Bits | VRAM @ 16K | Quality |
|---|---|---|---|
| Q3_K_M | 4 | 12.7 GB 12.5 + 0.2 KV | low |
| Q3_K_L | 4.3 | 13.6 GB 13.4 + 0.2 KV | moderate |
| IQ4_XS | 4.46 | 14.1 GB 13.9 + 0.2 KV | moderate |
| Q4_K_S | 4.67 | 14.7 GB 14.5 + 0.2 KV | moderate |
| Q4_K_M | 4.89 | 15.4 GB 15.2 + 0.2 KV | good |
| Q5_K_S | 5.57 | 17.4 GB 17.2 + 0.2 KV | good |
| Q5_K_M | 5.7 | 17.8 GB 17.6 + 0.2 KV | good |
| Q6_K | 6.56 | 20.4 GB 20.2 + 0.2 KV | excellent |
| Q8_0 | 8.5 | 26.2 GB 26.0 + 0.2 KV | lossless |
| FP16 | 16 | 48.7 GB 48.5 + 0.2 KV | lossless |
Select your GPU above to see speed estimates and compatibility for each quantization.
Rent a GPU and run LFM2 24B A2B now
Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.
Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.
At Q4_K_M quantization. Sorted by minimum VRAM.
Find the best GPU for LFM2 24B A2B
Build Hardware for LFM2 24B A2BRun LFM2 24B A2B locally with Ollama — needs 15.2 GB VRAM at Q4_K_M:
ollama run lfm2:24b