TextCompressor is a drop-in proxy that compresses prompts before they reach your LLM β removing stop words and filler while
preserving meaning.
Point your existing OpenAI client at our API, add one header, done.
β Light: 16.7% token savings, -2.7pp accuracy
β Medium: 33.5% token savings, -5.1pp accuracy
β Aggressive: 45.9% token savings, -6.6pp accuracy
Works with OpenAI, Anthropic, Ollama, LM Studio β anything OpenAI-compatible. No AI used in compression β pure CPU
