LFM2.5-2.6B
2recorded runs
305.6best tok/s
2verified
Fastest GPUs for this model (best tok/s)
- RTX 3090305.6
- Intel Arc Pro B70132.4
| GPU | Model | Quant | Backend | Ctx | tok/s | prompt tok/s | VRAM GB | Source |
|---|---|---|---|---|---|---|---|---|
| RTX 3090 · 24 GB | LFM2.5-2.6B | Q4_K_M | llama.cpp 0.3.0-dev (build 1, commit d7bd3bf) | 24576 | 305.6 ✓ | 7609.8 | 2.3 | src |
| Intel Arc Pro B70 · 32 GB | LFM2.5-2.6B | Q8_0 | llama.cpp 9fee29e9435f; oneAPI 2026.1.1 | 8192 | 132.4 ✓ | — | — | src |