Last released Jun 15, 2026
Cut your LLM prompt size by 40-70% in one line of code -- semantic chunking + extractive summarization that preserves meaning, instructions, and key entities.
Supported by
