supreme-draft-0.5b is the ultra-fast 0.5B Speculative Draft Model for SupremeAI 2.0. Designed to accelerate inference speed by up to 2.5x via speculative decoding. Merged via DARE-TIES using Qwen2.5-0.5B-Instruct and Qwen2.5-0.5B.
Qwen2.5-0.5B-Instruct
Qwen2.5-0.5B
Chat template
Files info
Base model
