Which API should you use with Kimi Code CLI?

For Kimi Code CLI on one monthly price with more than Kimi models, the best API is a flat-rate OpenAI-compatible provider: add Standard Compute to ~/.kimi-code/config.toml as an openai provider at https://api.stdcmpt.com/v1 and pin moonshotai/kimi-k3 or let routing pick, from $19 a month with a $20 compute budget. A Kimi membership is the native route: Plus at $19 a month and up include Kimi Code, with a rolling 5-hour window and a monthly total. The old Python Kimi CLI is archived; Kimi Code CLI (the kimi command) replaced it.

Guide updated 2026-10-14. Check each provider's current terms below.
Light or occasional use
If you only use Kimi models and stay inside the 5-hour windows, Kimi Plus at $19 is hard to beat. For a few sessions a month, a pay-as-you-go Moonshot API key will likely cost less.
One fixed monthly bill
Standard Compute fits when you want Kimi K3 next to Claude, GPT or GLM on one key, without 5-hour windows. Standard Compute draws Kimi K3 from the budget at $3 / $15 per million tokens; requests stop at the monthly budget instead of billing overage.

API and subscription options for Kimi Code

OptionPriceWhat you get
Kimi membership$19 / $39 / $99 / $199 per moKimi Code is included from Plus upward (Plus, Pro, Max, Ultra), with a rolling 5-hour window plus a monthly total and no weekly cap. Kimi models only. USD prices for the international site. Source: kimi.com
Standard Compute (us)From $19/moSmart routing sends each request to the cheapest model that can do it, or pin any model by its exact ID. Standard Compute starts at $19/month with a $20 monthly compute budget. Requests stop when that budget is exhausted until renewal or a plan change; there are no automatic overage charges. Requests are not slowed as the budget is used. It is a fixed bill with a finite allowance.
Moonshot (Kimi Platform) APIPay-as-you-goKimi models billed per token with a Kimi Platform key, set up as Kimi Code's built-in kimi provider. Suits light or bursty use. Source: platform.moonshot.ai
OpenRouterPay-as-you-goChoose models and providers through one API. Usage-based pricing can suit light or bursty work; estimate long agent sessions and review current fees and limits. Source: openrouter.ai

Prices and allowances can change. Kimi Code's provider format and membership terms come from Kimi Code's docs, read on October 14, 2026; membership prices are the international USD prices shown on kimi.com that day. Kimi Code: providers and models · Kimi Code membership · Kimi CLI (archived). Spot an error? Tell us.

Add Standard Compute to Kimi Code CLI

~/.kimi-code/config.toml
default_model = "standardcompute"

[providers.standardcompute]
type = "openai"
base_url = "https://api.stdcmpt.com/v1"
api_key_env = "STANDARDCOMPUTE_API_KEY"

[models.standardcompute]
provider = "standardcompute"
model = "standardcompute"
max_context_size = 200000
capabilities = ["tool_use"]

[models.kimi-k3]
provider = "standardcompute"
model = "moonshotai/kimi-k3"
max_context_size = 1048576
capabilities = ["tool_use", "thinking", "image_in"]

Export STANDARDCOMPUTE_API_KEY and run kimi; switch models with -m kimi-k3. Kimi Code only reads the key from the variable named in api_key_env, never from other shell variables.

Kimi Code questions

Does Kimi CLI still work, or do I need Kimi Code?
Move to Kimi Code CLI. Moonshot archived the Python Kimi CLI (MoonshotAI/kimi-cli); its README says it has been replaced by Kimi Code CLI and that existing installations will stop working. Kimi Code installs with npm i -g @moonshot-ai/kimi-code or Homebrew, keeps the kimi command, and reads ~/.kimi-code/config.toml.
Can I use Kimi K3 in Kimi Code without a Kimi membership?
Yes. Kimi Code CLI connects to any OpenAI Chat Completions provider through a type = "openai" entry, so a Standard Compute key serves moonshotai/kimi-k3, plus Claude, GPT and GLM, from $19 a month with a $20 compute budget. Declare capabilities such as tool_use on the model, since Kimi Code matches capabilities by model name and may not recognise a gateway's IDs. Kimi's SearchWeb tool works only on Kimi's own platform.

Try one task before choosing a plan

  1. Check the API format and model configuration supported by your installed Kimi Code version.
  2. Run a small task on a clean branch and review the result, elapsed time and compute used.
  3. Estimate your actual monthly workload before selecting a larger allowance.
Get your API keyKimi Code setup guide →Compare compute budgets →

Record your own results with the workload CSV and report template. These are blank templates, not published benchmark results.

Disclosure: this guide is maintained by Standard Compute, one of the options above. Choose based on task quality, supported features and total cost for your workload. Light usage may cost less on a usage-based API; sustained workloads need enough allowance on any subscription.

All flat-rate plans compared →The cheapest-tokens math →Plan compute for your team →