1 posts
0 posts
Hugging Face
Models, datasets, and tooling from the Hugging Face ecosystem.
NVIDIA cut Qwen3.6-35B 3×. Accuracy barely moved.
NVIDIA's NVFP4 Qwen3.6-35B checkpoint on HuggingFace: 3.06× memory reduction, <1% accuracy loss, Blackwell-native, vLLM flags included.
이 섹션에 아직 한국어 글이 없습니다.