community pool / model answer

OpenReasoning-Nemotron-32B

chat · nvidia/OpenReasoning-Nemotron-32B

21.8 tok/s · model median · 6 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
RTX 3090 24GB4-bit21.8 tok/smeasured-0 GB65,536llama.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.