community pool / model answer
Mellum2-12B-A2.5B-Thinking-GGUF-Q4_K_M
chat · JetBrains/Mellum2-12B-A2.5B-Thinking-GGUF-Q4_K_M
188.9 tok/s · model median · 2 runs in the model record.
| Reference rig | Quantization | Median result | Basis | TTFT (ms) | Peak VRAM | Max context | Engines |
|---|---|---|---|---|---|---|---|
| M5 Pro 24GB | 4-bit | 112.3 tok/s | reported | 283 | - GB | 4,224 | llama.cpp |
| RTX 5070 12GB | 4-bit | 265.5 tok/s | reported | 67 | - GB | 2,048 | 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.