Falcon3-7B-Instruct
1recorded run
19.9best tok/s
0verified
Fastest GPUs for this model (best tok/s)
- Tenstorrent P15019.9
| GPU | Model | Quant | Backend | Ctx | tok/s | prompt tok/s | VRAM GB | Source |
|---|---|---|---|---|---|---|---|---|
| Tenstorrent P150 · 32 GB | Falcon3-7B-Instruct | bfp_bf8 | vllm unknown | 2048 | 19.9 … | 1821.4 | — | src |