One API. All Modalities. Built for Asia. OpenAI-compatible inference for open models — LLM, vision, image, speech, embeddings.

Visit our docs to build custom applications.