amd / PARD2-Qwen3-8B

Text Generation752M2.7K0Updated Jun 15, 2026

Downloads

2.7K

Likes

0

Last updated

Jun 15, 2026

License

mit

PARD is a high-performance speculative decoding method that also enables low-cost adaptation of autoregressive draft models into parallel draft models. PARD-2 further advances PARD by introducing a Target-Aligned Parallel Draft Model for dual-mode speculative decoding. Instead…

Tasks & library

  • text generation
  • transformers

Tags

  • transformers
  • qwen3
  • text-generation

Technical specifications

Parameters752M
Librarytransformers
LanguagesNot available
Licensemit