macrostack

AI API price index · model maker · read 17 Nov 2026

Moonshot AI (Kimi) API pricing

11 prices as Moonshot AI (Kimi) publishes them — 6 text models, 5 search and fetch — read on its own pricing page on 17 Nov 2026. Its current text models cost $10.00 to $15.00 per million output tokens.

Free tier No free tier: "a minimum recharge of $1 is required to start using the API; when your cumulative recharge reaches $5, you receive a $5 voucher." Rate limits rise with the cumulative recharge (Tier0 at $1: 3 requests a minute; Tier1 at $10: 100).

What to know before you build on Moonshot AI (Kimi)

  • Kimi K3.1 Pro is $2.00 in and $10.00 out per million tokens, Kimi K3 $3.00 and $15.00, both with a 1,048,576-token context. Cached input is a tenth of the input price.
  • The highspeed variants are a faster serving path at twice the price: kimi-k3.1-pro-highspeed is $4.00 in and $20.00 out.
  • The Batch API costs 60% of the standard price on Kimi K3.1 Pro and Kimi K3.
  • The K2 series (kimi-k2.7-code, kimi-k2.6) closed to accounts registered from 31 October 2026 and is to be retired by the end of December 2026.
  • There is no free tier: the first recharge is at least $1, and a $5 voucher arrives once $5 has been recharged.
  • Web search is billed per call on top of tokens: $0.002 for Basic, $0.003 for Pro, $0.002 for a URL fetch.

Against the other 31

Across the index, 390 current text models run from $0.03 to $50.00 per million output tokens — 1667×. Moonshot AI (Kimi)’s cheapest, kimi-k3.1-pro at $10.00, has 347 cheaper text models elsewhere; its dearest, kimi-k3, costs $15.00. A cheaper model is a different model, not the same one for less — compare capability before price.

The same model elsewhere

Open-weight models Moonshot AI (Kimi) sells that other hosts price too — the only like-for-like comparison in the index.

  • Kimi K3 — DeepInfra $2.85 / $14.25 · Amazon Bedrock $3.00 / $15.00 · Fireworks AI $3.00 / $15.00 · Moonshot AI (Kimi) $3.00 / $15.00 · Nebius Token Factory $3.00 / $15.00 · Novita AI $3.00 / $15.00 · Perplexity $3.00 / $15.00 · Together AI $3.00 / $15.00 — 1.1× apart
  • Kimi K3.1 Pro — DeepInfra $2.00 / $10.00 · Moonshot AI (Kimi) $2.00 / $10.00 · Nebius Token Factory $2.00 / $10.00 · Novita AI $2.00 / $10.00 — the same everywhere

Input / output per 1M tokens, standard tier.

Text models

ModelInputCached inputOutputNote
kimi-k3.1-pro$2.00$0.20$10.00K3 series, 1,048,576-token context; cache writes $2.00 (5m) and $4.00 (1h) per 1M tokens
kimi-k3.1-pro-highspeed$4.00$0.20$20.00fast serving path of kimi-k3.1-pro; K3 series, 1,048,576-token context; cache writes $4.00 (5m) and $8.00 (1h) per 1M tokens · Not ranked: faster serving path of a model ranked under its default name.
kimi-k3$3.00$0.30$15.00K3 series, 1,048,576-token context; cache writes $3.00 (5m) and $6.00 (1h) per 1M tokens
kimi-k2.7-codelegacy$0.95$0.19$4.00K2 series, 262,144-token context; closed to accounts registered from 31 October 2026, to be retired by the end of December 2026 · Not ranked: legacy model, existing customers only.
kimi-k2.7-code-highspeedlegacy$1.90$0.38$8.00K2 series, 262,144-token context; closed to accounts registered from 31 October 2026, to be retired by the end of December 2026 · Not ranked: legacy model, existing customers only.
kimi-k2.6legacy$0.95$0.16$4.00K2 series, 262,144-token context; closed to accounts registered from 31 October 2026, to be retired by the end of December 2026 · Not ranked: legacy model, existing customers only.

Search and fetch

ModelPriceNote
Web Search Basic$0.002 per requestcharged only when the call succeeds and returns results or content
Web Search Pro$0.003 per requestcharged only when the call succeeds and returns results or content
URL Fetch$0.002 per requestcharged only when the call succeeds and returns results or content
Responses API web_search tool$0.005 per requestcharged per search executed; model tokens are billed on top
Formula API web-search tool$0.005 per requestcharged per search executed; model tokens are billed on top

2 of these 11 prices are ranked in the index; the rest say why in their note. Text and embeddings per 1M tokens unless the row says otherwise.

Discounts and surcharges, as published

  • Batch API: "Batch API inference costs 60% of the standard model price", on kimi-k3.1-pro and kimi-k3; cache hits are best effort in batch.
  • Context caching is on by default (TTL 5 minutes): cached input costs a tenth of the input price on kimi-k3 and kimi-k3.1-pro. K3 cache writes are billed by TTL, at the input price for 5 minutes and twice it for an hour.
  • Prices exclude applicable taxes, "calculated at checkout based on your jurisdiction".

Where these numbers come from

Read on platform.kimi.ai/docs/pricing/chat on 17 Nov 2026, checked against 3 more of Moonshot AI (Kimi)’s own pages. Macrostack measures nothing and is paid by no provider on this index. Prices change; the date is the claim.

The other providers in the index

Questions

How much does the Moonshot AI (Kimi) API cost?
On 17 Nov 2026, Moonshot AI (Kimi)'s current text models cost $10.00 to $15.00 per million output tokens at the standard tier (kimi-k3.1-pro to kimi-k3). The full list, with input and cached-input prices, is on this page.
Does Moonshot AI (Kimi) have a free tier?
What Moonshot AI (Kimi)'s own pages say: No free tier: "a minimum recharge of $1 is required to start using the API; when your cumulative recharge reaches $5, you receive a $5 voucher." Rate limits rise with the cumulative recharge (Tier0 at $1: 3 requests a minute; Tier1 at $10: 100).
Does Moonshot AI (Kimi) offer discounts?
Batch API: "Batch API inference costs 60% of the standard model price", on kimi-k3.1-pro and kimi-k3; cache hits are best effort in batch. Context caching is on by default (TTL 5 minutes): cached input costs a tenth of the input price on kimi-k3 and kimi-k3.1-pro. K3 cache writes are billed by TTL, at the input price for 5 minutes and twice it for an hour. Prices exclude applicable taxes, "calculated at checkout based on your jurisdiction".
The Macrostack brief

New swaps, worth your inbox.

A short, occasional email when we add a high-intent alternative or ship a new head-to-head. No spam, no selling your address — unsubscribe in one click.