community pool / model answer
MiniMax-M2.7
chat · unsloth/MiniMax-M2.7
14.3 tok/s · model median · 6 runs in the model record.
| Reference rig | Quantization | Median result | Basis | TTFT (ms) | Peak VRAM | Max context | Engines |
|---|---|---|---|---|---|---|---|
| Arc Pro B70 32GB ×4 | 4-bit | 14.1 tok/s | measured | - | 26.4 GB | 64 | llama.cpp |
| GB10 Grace Blackwell 128GB | 4-bit | 30.9 tok/s | reported | 795 | 112 GB | 108,000 | 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.