Paired-4:8 + NVFP4 W4A4 expert compressed MoE models for NVIDIA Blackwell Sparse Tensor Cores.
-
ISTA-DASLab/Kimi-K2.5-P48NVFP4-MoESQ
Text Generation • 646B • Updated • 487 -
ISTA-DASLab/Qwen3-30B-A3B-P48NVFP4-MoESQ
Text Generation • 20B • Updated • 618 • 3 -
ISTA-DASLab/Qwen3.5-397B-A17B-P48NVFP4-MoESQ
Image-Text-to-Text • 258B • Updated • 67 -
ISTA-DASLab/Qwen3.8-Flash-Next-P48NVFP4-MoESQ
Text Generation • 135B • Updated • 870 • 9