community pool / model answer

Muse-Glimmer-30B-NVFP4

chat · Preyazz/Muse-Glimmer-30B-NVFP4

124.8 tok/s · model median · 3 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
RTX 5090 32GB4-bit94.1 tok/sreported31- GB8,192vllm
RTX PRO 6000 Blackwell 96GB4-bit152.2 tok/sreported22- GB32,768vllm

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.