The quickstart lists two model IDs: kimi-k3.1-pro and kimi-k3.1-pro-highspeed, the latter described as delivering two to three times the speed of the base Pro model.
The documentation specifies a one-million-token context window, vision input, tool calling, and streaming responses with reasoning content separated from the main content. Thinking is always enabled. A top-level reasoning_effort parameter accepts low, high, or max, with max as the default.
API access is unlocked after a successful top-up of at least $1. Cumulative top-ups determine the account tier and rate limits. Moonshot describes K3.1 Pro as a post-trained Kimi K3 model and says full weights will be released by November 4.
The source does not provide API token pricing, current rate-limit figures, a release year for the weights date, or independently verified benchmark results. Several benchmark values displayed in the documentation excerpt are placeholders.