WebWorld-8B
1recorded run
62.3best tok/s
1verified
Fastest GPUs for this model (best tok/s)
- RTX 3060 x262.3
| GPU | Model | Quant | Backend | Ctx | tok/s | prompt tok/s | VRAM GB | Source |
|---|---|---|---|---|---|---|---|---|
| RTX 3060 x2 · 12 GB | WebWorld-8B | Q4_K_M | llama.cpp 2d97363 | 512 | 62.3 ✓ | 1976.2 | 6.3 | src |