Threads @iammurataslann — Ddz0PfsiMeQ
Murat Aslan (@iammurataslann) · Threads · source
NVIDIA Drop 63% smaller Qwen3.8-Flash-Next NVFP4 quantized on Hugging Face
125B MoE with hybrid attention, now 63% smaller with minimal accuracy loss.
Murat Aslan (@iammurataslann) · Threads · source
NVIDIA Drop 63% smaller Qwen3.8-Flash-Next NVFP4 quantized on Hugging Face
125B MoE with hybrid attention, now 63% smaller with minimal accuracy loss.