community pool / model answer
gemma-4-12B-it-qat-GGUF
chat · unsloth/gemma-4-12B-it-qat-GGUF
79 tok/s · model median · 12 runs in the model record.
| Reference rig | Quantization | Median result | Basis | TTFT (ms) | Peak VRAM | Max context | Engines |
|---|---|---|---|---|---|---|---|
| M5 Pro 24GB | 4-bit | 67.2 tok/s | reported | 250 | - GB | 2,048 | llama.cpp |
| RTX 3060 12GB | 4-bit | 79 tok/s | reported | 1,245 | 9.5 GB | 65,536 | llama.cpp |
| RTX 3090 24GB | 4-bit | 104.1 tok/s | measured | - | 11.8 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.