community pool / model answer

Mellum2-12B-A2.5B-Instruct

chat · JetBrains/Mellum2-12B-A2.5B-Instruct

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

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
RX 570 4GB ×24-bit62.7 tok/sreported95- GB2,048llama.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.