Last released May 18, 2026
Cross-platform text compression that uses a local GPT-2 (via llama.cpp) as the probability model.
Supported by
