Kimi K2 Thinking pricing
Moonshot AI ·
moonshot-kimi-k2-thinking ·
262K context
Estimate Kimi K2 Thinking for your workload Source and check date
Rates last checked against LiteLLM’s community-maintained price map on 2026-09-22.
What Kimi K2 Thinking costs
| Rate | Per 1M tokens | Per 1K tokens | What it covers |
|---|---|---|---|
| Input | $0.6 | $0.0006 | Every token you send: prompt, history, documents. |
| Output | $2.5 | $0.0025 | Every token the model generates, including hidden reasoning tokens. |
| Cached input | $0.15 | $0.00015 | Input served from the provider’s prompt cache. |
What that means in practice
The same three workloads are costed on every model page, so these numbers are directly comparable across the catalog. No prompt caching and no batch discount is assumed — the figures are what you pay before you optimize anything.
Support chatbot
2,500 six-turn conversations a day
$2,214 / month
$0.030 per conversation · $26,568 a year
An 800-token system prompt, 400-token questions and 900-token answers, over six turns — each turn re-sending everything before it.
Document summarizer
1,000 single-shot summaries a day
$280 / month
$0.0093 per conversation · $3,355 a year
One 12,000-token document in, an 800-token summary out, no conversation history. The case where input dominates the bill.
Coding agent
200 twelve-turn sessions a day
$1,595 / month
$0.266 per conversation · $19,138 a year
A 3,000-token system prompt, 1,500-token instructions and 2,500-token responses over twelve turns. Long sessions are where the quadratic cost of re-sent history stops being theoretical.
Change the assumptions in the calculator
Cheaper models in the catalog
Nothing cheaper is scored as highly as Kimi K2 Thinking, so these are simply the cheaper options, best-scored first. Trading down here is a real trade.
| Model | Input | Output | Blended saving | |
|---|---|---|---|---|
| DeepSeek V4 Flash | $0.3 | $1.2 | 51% | |
| MiMo V2.6 Pro | $0.435 | $0.87 | 49% | |
| Claude Haiku 5.5 | $0.1 | $0.5 | 81% | |
| GLM-5.3 Flash | $0.15 | $0.5 | 78% | |
| MiMo V2.6 Flash | $0.14 | $0.28 | 84% |
Specification
| Context window | 262K tokens |
|---|---|
| Maximum output | 262K tokens |
| Provider | Moonshot AI |
| Status | Retired by its vendor |
| Reasoning model | Yes |
| Vision | Yes |
| Token counting | Estimated at ~3.4 characters per token (Estimate — Moonshot tokenizer, calibrated for mixed English/Chinese) |
Price history
PromptSpend has recorded Kimi K2 Thinking since 2026-08-02, and none of its rates has changed in that time. Tracking began on 2026-08-02, so anything the vendor did before then is not on record here.
Every row, with the commit that published it, is in the price-history dataset; price news is also available as an Atom feed.
Where these numbers come from
Last checked against its source on 2026-09-22. Source of record: LiteLLM’s community-maintained price map.
Moonshot AI’s own pricing page
Regional and data-residency premiums, priority tiers, server-side tool fees and negotiated discounts are not included. Prices change without notice; this page is rebuilt every time the catalog does.
Common questions
How much does Kimi K2 Thinking cost?
Kimi K2 Thinking costs $0.6 per 1M input tokens and $2.5 per 1M output tokens at the standard tier, as last checked against LiteLLM’s community-maintained price map on 2026-09-22.
How much does Kimi K2 Thinking cost to run for a month?
On PromptSpend's three standard workloads, Kimi K2 Thinking costs about $2,214 for a support chatbot, $280 for a document summarizer and $1,595 for a coding agent each month. No prompt caching and no batch discount is assumed, and the same workloads are costed on every model page, so the figures compare directly.
Is there a cheaper alternative to Kimi K2 Thinking?
Yes, though none is scored as highly. The cheaper options, best-scored first, include DeepSeek V4 Flash ($0.3 in, $1.2 out), MiMo V2.6 Pro ($0.435 in, $0.87 out) and Claude Haiku 5.5 ($0.1 in, $0.5 out), so trading down is a real trade.
How large is the context window of Kimi K2 Thinking?
Kimi K2 Thinking accepts up to 262K tokens of context and can produce up to 262K tokens of output.
Has the price of Kimi K2 Thinking changed?
No vendor change is on record. PromptSpend has recorded Kimi K2 Thinking since 2026-08-02 and none of its rates has moved. PromptSpend began tracking prices on 2026-08-02, so earlier changes are not on record.
Where do the prices for Kimi K2 Thinking come from?
They come from LiteLLM’s community-maintained price map and were last checked on 2026-09-22. Regional and data-residency premiums, priority tiers, server-side tool fees and negotiated discounts are not included.