community pool / model answer

gemma-4-26b-a4b-it-4bit

chat · mlx-community/gemma-4-26b-a4b-it-4bit

106.9 tok/s · model median · 2 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
M3 Ultra 512GB4-bit96.7 tok/sreported145- GB8,192mlx
Max 128GB4-bit117 tok/sreported-- GB2,048mlx

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.