High-speed inference variant of Kimi's most intelligent coding model with a native multimodal architecture supporting text, image, and video input, thinking modes, and agent tasks.
LLM Gateway routes requests to the best providers that are able to handle your prompt size and parameters.