Bonsai-27B
Qwen family · 28.0B params
2recorded runs
136.7best tok/s
1verified
All benchmarks Methodology
| GPU | Model | Quant | Backend | Ctx |
tok/s | prompt tok/s | VRAM GB | Source |
| RTX 3080 · 10 GB | Bonsai-27B | Q1_0 | llama.cpp b10472 | 32768 | 72.9 ✓ | 1319.8 | 5.1 | src |
| RTX 5090 · 32 GB | Bonsai-27B | Q1_0 | lmstudio unknown | 2048 | 136.7 … | 389.4 | — | src |