community pool / model answer
Qwen3.5-122B-A10B
chat · Qwen/Qwen3.5-122B-A10B
25.4 tok/s · model median · 10 runs in the model record.
| Reference rig | Quantization | Median result | Basis | TTFT (ms) | Peak VRAM | Max context | Engines |
|---|---|---|---|---|---|---|---|
| GB10 Grace Blackwell 128GB | 6-bit | 17.1 tok/s | reported | - | - GB | 100,000 | llama.cpp |
| GTX 1080 Ti 11GB | 3-bit | 3.2 tok/s | reported | - | 10.3 GB | 8,192 | llama.cpp |
| Radeon AI Pro R9700 32GB ×3 | 4-bit | 26.5 tok/s | measured | 1,296 | 77.1 GB | 65,536 | llama.cpp |
| Ryzen AI Max 395 128GB | 4-bit | 18 tok/s | reported | 3,418 | 74 GB | 262,144 | llama.cpp |
Context tested
The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.