community pool / model answer
Qwen3.5-35B-A3B
chat · Qwen/Qwen3.5-35B-A3B
85.6 tok/s · model median · 7 runs in the model record.
| Reference rig | Quantization | Median result | Basis | TTFT (ms) | Peak VRAM | Max context | Engines |
|---|---|---|---|---|---|---|---|
| Arc Pro B70 32GB | 4-bit | 62.8 tok/s | reported | - | - GB | 262,144 | llama.cpp |
| Arc Pro B70 32GB | 5-bit | 61.5 tok/s | reported | - | - GB | 262,144 | llama.cpp |
| GTX 1080 Ti 11GB | 4-bit | 9.3 tok/s | reported | - | 9.4 GB | 8,192 | llama.cpp |
| M5 Max 128GB | 4-bit | 122 tok/s | reported | - | - GB | 262,144 | ollama |
| RX 7900 XTX 24GB | 4-bit | 140.6 tok/s | measured | 75 | 22.1 GB | 4,096 | hipfire |
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.