The Kimi Code CLI is free; access to the models behind it is sold in three ways — a Kimi membership starting at $19 per month, an Extra Usage wallet that takes over when your membership quota runs out, and the Kimi Open Platform pay-as-you-go API, priced at $3 per million input tokens and $15 per million output tokens for K3. The membership quota is not published, so choosing among the three depends on usage. The calculation below shows where the break-even point falls. Prices read from kimi.com and platform.kimi.ai on October 1, 2026.
To understand what Kimi Code is and how it compares, see Kimi Code vs Claude Code; to switch clients, memberships, or models, see Kimi Code alternatives. The English original of this page is Kimi Code pricing.
Membership plans
| Plan | Monthly | Annual plan | Kimi Code and K3 | K3 1M context, HighSpeed |
|---|---|---|---|---|
| Plus | $19 | $15 per month ($180 per year) | Yes | No |
| Pro | $39 | $31 per month ($372 per year) | Yes | Yes |
| Max | $99 | $79 per month ($948 per year) | Yes | Yes |
| Ultra | $199 | $159 per month ($1.908 per year) | Yes | Yes |
Prices are from kimi.com's membership page, where they are rendered in the browser rather than in the page source; the two columns on the right are from Kimi's membership documentation. Free accounts and the Go plan do not include Kimi Code quota. Kimi recently renamed the plans while saying “prices unchanged”; existing plans keep their old names and rules until you switch. All amounts are in US dollars; the actual amount charged in Brazilian reais depends on the exchange rate and IOF applied by your card.
How the quota works
- Everything shares one pool. Requests from the CLI, VS Code, the desktop app, and third-party tools all count against the same membership quota, shared by all signed-in devices and API keys.
- A rolling five-hour window limits bursts and recovers automatically. New subscribers no longer have a weekly limit; legacy plans still renew every seven days from the subscription date, with no rollover.
- A total monthly quota, shared with Kimi on the web, freezes Kimi Code when reached, until reset or upgrade.
- No figures are published for any of the three — neither tokens nor requests. Remaining quota appears in the Kimi Code Console and in the CLI with
/usage.
Extra Usage: the wallet beyond your quota
When the membership quota runs out, Extra Usage continues requests from a prepaid balance. The rules, from the same documentation:
- Only active subscribers can enable it, and Kimi on the web and Kimi Code use one shared balance.
- Billed in yuan according to actual usage, at rates Kimi describes as close to Open Platform API prices — Kimi itself says that upgrading the subscription is usually better value for heavy use.
- It is deducted last, after the subscription quota, so tasks keep running instead of failing. Without a monthly spending limit configured, there is no limit.
- Top-ups start at ¥25, with a daily maximum of 10 top-ups and ¥3,000, and a balance cap of ¥10,000. The balance does not expire, remains if the subscription lapses (but you cannot top up while it is lapsed), and is generally non-refundable.
Pay-as-you-go API pricing per token
| Model ID | Input (cache miss) | Input (cache hit) | Output | Context |
|---|---|---|---|---|
kimi-k3 — escrita em cache $3,00 (TTL de 5 minutos) ou $6,00 (1 hora) | $3,00 | $0,30 | $15,00 | 1.048.576 tokens |
kimi-k2.7-code | $0,95 | $0,19 | $4,00 | 262.144 tokens |
kimi-k2.7-code-highspeed | $1,90 | $0,38 | $8,00 | 262.144 tokens |
Per million tokens, before tax, from platform.kimi.ai; the API base URL is https://api.moonshot.ai/v1 and accepts requests in OpenAI and Anthropic formats. This is a separate product from the membership: its own account and key, and different model IDs (kimi-k3, kimi-k2.7-code) from those used by the membership (kimi-for-coding, k3). The membership cannot pay for API calls, and API balance does not increase membership quota.
Membership or pay-as-you-go: where is the break-even point?
The following is token arithmetic, not measured task cost. One session sends 200.000 uncached input tokens and receives 12.000 output tokens:
| Route | Session | Sessions equivalent to Plus ($19) | Who bills you |
|---|---|---|---|
kimi-k3, pay-as-you-go | $0,78 | about 24 | Kimi Open Platform |
kimi-k3, 80% input cache hit | $0,35 | about 55 | Kimi Open Platform |
kimi-k2.7-code, pay-as-you-go | $0,24 | about 80 | Kimi Open Platform |
| Claude Haiku 4.5 through Kunavo | $0,18 | about 104 | Kunavo |
| Claude Sonnet 5 through Kunavo | $0,36 | about 52 | Kunavo |
Read this as a threshold. If you run fewer K3 sessions of this kind per month than the number in the third column, the pay-as-you-go API costs less than Plus; if you run more, Plus is probably better value — until the five-hour window or monthly total quota stops you, which is precisely what Kimi has not quantified. The cached row excludes the first cache-write charge, which is $3 or $6 per million depending on cache lifetime. “Lowest unit price” and “cheapest way to finish a task” are different questions: a model that needs three retries may cost more than a more expensive model that finishes in one.
Kunavo’s rates are live catalog rates for Claude models, which Kimi Code accesses through the openai provider type; Kunavo does not sell Kimi models. The catalog price is a billing floor, not a ceiling—when the original provider reports a cost, billing is the greater of the catalog cost and the source cost multiplied by the applicable margin. Kunavo charges per token from a prepaid balance, with a minimum top-up of $10 and no subscription (billing); in Stripe Checkout, people paying from Brazil see Pix, with the amount converted to reais before confirmation, as well as card, Apple Pay, Google Pay, and Link. The provider block is on the Kimi Code configuration page in Kunavo. No one at Kunavo has run Kimi Code against this endpoint, so check the charge for the first request in your usage log.
Frequently asked questions
How much does Kimi Code cost?
The Kimi Code CLI is free software under the MIT license (MoonshotAI/kimi-code, version 2.1.1 dated September 24, 2026). What you pay for is model access, sold in three ways. A Kimi membership starting at Plus includes Kimi Code: as of October 1, 2026, kimi.com listed Plus at $19 per month ($15 per month on an annual plan, $180), Pro at $39 ($31, $372), Max at $99 ($79, $948), and Ultra at $199 ($159, $1,908). Extra Usage, a separate wallet in yuan, continues when the membership quota runs out. Or you can skip the membership and pay per token on the Kimi Open Platform, where kimi-k3 costs $3 per million input tokens and $15 per million output tokens.
Which Kimi plan do I need for Kimi Code and K3?
Plus is the entry point: Kimi's membership documentation says Kimi Code and K3 access are available from Plus and up, while K3's 1M-token context and HighSpeed access require Pro or higher. The Go plan has no coding quota. Higher-tier plans increase the agent quota on kimi.com — the pricing page shows 2x for Pro, 5x for Max, and 10x for Ultra compared with Plus — but no plan publishes a token or request count for Kimi Code.
What happens when Kimi Code quota runs out?
New subscribers have two limits. A rolling five-hour window limits bursts and recovers automatically; the weekly limit has been removed for new subscribers, though legacy plans still reset every seven days. A total monthly quota, shared with Kimi on the web, freezes Kimi Code until it resets or you upgrade. After that, Extra Usage continues requests from a prepaid balance in yuan, billed at rates Kimi says are close to Open Platform API prices. With Extra Usage enabled and no monthly spending limit configured, there is no limit.
Is a Kimi membership cheaper than the Kimi API?
For heavy use, probably — Kimi's own documentation says upgrading usually offers better value than using a lot of Extra Usage. The break-even point depends on volume. At the Open Platform rate for kimi-k3, a session that sends 200,000 uncached input tokens and receives 12,000 output tokens costs about $0.78, so $19 for Plus equals about 24 such sessions per month; with kimi-k2.7-code, about 80. Kimi does not publish how many sessions Plus actually allows, so the honest test is: if your metered API use for a month would cost more than the plan, try the plan and track usage in the Kimi Code Console.
Can I use another API, or a cheaper one, with Kimi Code?
Yes. The Kimi Code provider documentation describes an openai provider type for any OpenAI-compatible service with a changed base_url, configured in ~/.kimi-code/config.toml. The /login selector offers only Kimi Code OAuth or a Kimi Platform API key, so other providers require editing the file. Membership and Open Platform use different model IDs and endpoints, so a configuration written for one will not work for the other. Kunavo does not sell Kimi models; its Kimi Code setup page points the openai type to Claude and GPT IDs.
Read on October 1, 2026: kimi.com/membership/pricing (rendered in the browser), Kimi Code membership documentation, the chat pricing and quickstart pages on platform.kimi.ai, and the GitHub API for MoonshotAI/kimi-code. No Kimi account was used, and nothing was purchased or run; all amounts are calculations based on published rates. Membership quotas are unpublished, so no per-token plan price is given.