RedHatAI/NVIDIA-Nemotron-3-Ultra-550B-A55B-speculator.dflash

Quantization vs RedHatAI/NVIDIA-Nemotron-3-Ultra-550B-A55B-FP8-dynamic. RedHatAI/NVIDIA-Nemotron-3-Ultra-550B-A55B-FP8-dynamic is a quantized version of the other (F8_E4M3/BF16).

On the Hub · Downloads on Model Pulse