Compressed prompts won
Compressed prompts outperformed uncompressed in a 268K-vote blind arena across models.
Read the notes →Labs waste your tokens.
We don’t.
Every filler token in your context is revenue for a lab. With metre-2 in front, filler never wastes the context window. Same answers, a fraction of the cost.
Drawings, long prompts
On long agent runs analyzing construction drawings, about 47K tokens saved per prompt.
Read the notes →Process raw LLM inputs
We build proprietary compression models to process raw text. Below 50ms inference with full determinism and cache safety.
Compression encodes information with fewer tokens by dropping statistical filler while keeping semantic intent and logical relationships — so the same window holds more meaning.
Research
Conversational context
Metre-2 improved CoQA accuracy from 93.3% to 95.3% while cutting tokens by 8.2%.
Safety classifiers
Cut safety classifier costs by 30% while preserving or improving F1.
Wrap your existing client
One line wraps your OpenAI or Anthropic client. Your existing code stays the same. Compression happens automatically.
pip install token-metre from token_metre import wrap client = wrap(OpenAI())
Stop paying for filler.
Get an API key and see the savings on your own prompts.