Orthrus - a chiennv Collection

chiennv 's Collections

Orthrus

updated May 14

Fast, lossless LLM inference via dual-view diffusion decoding.

chiennv/Orthrus-Qwen3-4B

Text Generation • 5B • Updated May 16 • 94 • 9
chiennv/Orthrus-Qwen3-8B

Text Generation • 10B • Updated May 16 • 293 • 26
chiennv/Orthrus-Qwen3-1.7B

Text Generation • 2B • Updated May 16 • 956 • 9
Orthrus: Memory-Efficient Parallel Token Generation via Dual-View Diffusion

Paper • 2605.12825 • Published May 12 • 12