
TurnkeyML aims to make it easy to leverage the tools within the ONNX ecosystem via "no-code CLIs and low-code APIs" with their Turnkey optimizer and Lemonade SDK for large language models (LLMs) with the OnnxRuntime GenAI (OGA).

With the release of TurnkeyML 6.0, they are introducing a new OpenAI-compatible server. TurnkeyML has replaced their previous "serve" tool with a OpenAI-compatible server and the developers are also pursuing Ollama compatibility too.
TurnkeyML 6.0 also adds support for Quark quantization in a new "quark" tool, benchmark tooling improvements, and other improvements. Downloads and more details on TurnkeyML 6.0 via onnx/turnkeyml on GitHub .