Kimi K3 is Moonshot AI's open-weights flagship: a 2.8-trillion-parameter Mixture-of-Experts model with native vision and a 1M-token context window.
Moonshot bills Kimi K3 as "the world's first open 3T-class model, designed for frontier intelligence across long-horizon coding, knowledge work, and reasoning." Each token activates 16 of its 896 experts. The architecture is built on Moonshot's Kimi Delta Attention, and the released weights are MXFP4 with MXFP8 activations. Full details are in Moonshot's announcement.
On a Shared Endpoint, you pay per token. The endpoint is OpenAI-compatible and already live: point your existing SDK at it and start sending requests.
A Dedicated Endpoint is your own deployment of Kimi K3, on an inference stack that Modal tunes and autoscales. Bring fine-tuned weights if you have them. Compute is billed by the second, and only while it runs.
modal endpoint create --model moonshotai/Kimi-K3