Last released May 17, 2026
A lightweight Model I/O normalization layer for OpenAI-compatible LLM calls.
Supported by