Kimi Coding Plan covers the ways to use Moonshot AI's Kimi models for coding: a Kimi membership, which includes Kimi Code from the Plus tier, and pay-as-you-go access through the Kimi API. In October 2026 Moonshot AI launched Kimi K3.1 Pro, a post-trained version of Kimi K3 with a 1-million-token context window, together with Kimi K3.1 Pro Highspeed, which Moonshot says runs 2-3x faster. Both are available on Kimi.ai, Kimi Work, Kimi Code and the Kimi API, and Moonshot says the full model weights will be released by November 4, 2026.
The membership tiers are Adagio (free), Plus at $19/month ($15/month billed annually, $180/year), Pro at $39/month ($31/month billed annually), Max at $99/month ($79/month billed annually) and Ultra at $199/month ($159/month billed annually). Kimi Code is available on Plus and above but not on Adagio. Pro adds 2x agent credits, access to K3.1 Pro Highspeed and Goal mode; Max adds 5x agent credits and up to 1M tokens of context for long conversations; Ultra adds 10x agent credits. Concurrent agent tasks range from one on Adagio to four on Max and Ultra.
The Kimi API is billed per token with no subscription. Per 1M tokens, kimi-k3.1-pro costs $2 input and $10 output ($0.20 cached input), kimi-k3.1-pro-highspeed $4 and $20, kimi-k3 $3 and $15, and kimi-k2.7-code and kimi-k2.6 $0.95 and $4. Context caching is applied automatically with a default 5-minute TTL, with a 1-hour TTL as an option; K3.1 Pro unlocks after a top-up of at least $1, and Batch API jobs for K2.7 Code and K2.6 cost 60% of the standard price. Starting October 31, 2026, K2 series models are no longer available to newly registered users, and Moonshot plans to retire them by the end of December 2026. Kimi's documentation covers using its models in Claude Code, Codex and OpenCode as well as in Kimi Code.
