Repositories
Lossless greedy speculative decoding for 4-bit Qwen3.8-27B on one RTX 3090, with the acceptance logic proven in Bend
Python
The fastest decode and prefill for 4-bit Qwen3.8-27B on one RTX 3090 that preserves the model's intelligence
Python
Turn one SVG into optimized SVG, PDF, favicon.ico, and PNG, WebP and JPEG from 16 to 4096 px
Rust
TypeScript