Scheduled for retirement on Mar 30, 2027. Still available at the prices below until then.
Context
262K
Max output
262K
Serving providers
10
Cheapest input
$0.60 /1M
On-demand listed rates, OpenRouter routed quotes and Vercel reseller quotes. Enterprise commitments (committed-use discounts, provisioned throughput, negotiated rates) can be materially lower.
Benchmarks
independent evaluations · source: Artificial Analysis and Hugging Face leaderboardsSpeed & latency
Output speed
160 t/s
tokens / second
Time to first token
1.34s
latency
Headline indices
Individual benchmarks · click a row for effort variants
| Benchmark | Best score | Effort | Cost / task | |
|---|---|---|---|---|
| CritPt | 3% | default | — | |
| EvalComputeProxyFrozen Aug 18, 2026 | 542.5 | default | — | |
| GPQA Diamond | 84% | default | — | |
| Humanity's Last Exam | 24% | default | — | |
| IFBench | 68% | default | — | |
| LiveCodeBench | 85% | default | — | |
| Long-Context Reasoning | 72% | default | — | |
| MMLU-ProFrozen Aug 18, 2026 | 85% | default | — | |
| Omniscience | -21.4 | default | — | |
| Omniscience Accuracy | 0.3 | default | — | |
| Omniscience Non Hallucination | 0.2 | default | — | |
| SciCodeArtificial Analysis did not report this score in its latest run on Oct 14, 2026 | 42% | — | — | |
| SWE-bench Verified | 71% | default | — | |
| TerminalBench Hard | 31% | default | — | |
| τ²-bench | 93% | default | — |
Artificial Analysis last published scores for this benchmark on Aug 18, 2026. Scores here are frozen at that date, and models released since are not on it.
Pricing by serving provider
standard tier · per 1M tokens · click a provider to expandInput pricing runs from $0.47 to $2.50 per 1M tokens across 10 providers, so the dearest route costs 432% more than the cheapest for the same model.
| Serving provider | Input /1M | Output /1M | Endpoints | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| VercelVercel reseller quote | $0.47 | $2.00 | 1 | |||||||||||||
Standardmoonshotai/kimi-k2-thinking
| ||||||||||||||||
| Azure | $0.60 | $2.50 | 1 | |||||||||||||
Standardazure_ai/kimi-k2-thinking
| ||||||||||||||||
| Baseten | $0.60 | $2.50 | 1 | |||||||||||||
Standardbaseten/moonshotai/Kimi-K2-Thinking
| ||||||||||||||||
| Amazon Bedrock | $0.60 | $2.50 | 1 | |||||||||||||
Standardbedrock/moonshotai.kimi-k2-thinking
| ||||||||||||||||
| Fireworks AI | $0.60 | $2.50 | 1 | |||||||||||||
Standardfireworks_ai/accounts/fireworks/models/kimi-k2-thinking
| ||||||||||||||||
| MoonshotCheapest | $0.60 | $2.50 | 1 | |||||||||||||
Standardmoonshot/kimi-k2-thinking
| ||||||||||||||||
| Novita | $0.60 | $2.50 | 1 | |||||||||||||
Standardnovita/moonshotai/kimi-k2-thinking
| ||||||||||||||||
| OpenRouter | $0.60 | $2.50 | 1 | |||||||||||||
Standardmoonshotai/kimi-k2-thinking
| ||||||||||||||||
| Crusoe is dearest at $2.50 / $2.50 | ||||||||||||||||
Price history
listed seller prices, input and output per 1M tokens Input Output
Cost calculator
estimate your monthly spend on this model$254
estimated / month
Model IDs
copy the exact identifier for your platformmoonshotai/kimi-k2-thinkingbaseten/moonshotai/Kimi-K2-Thinkingfireworks_ai/accounts/fireworks/models/kimi-k2-thinkingmoonshot/kimi-k2-thinkingnovita/moonshotai/kimi-k2-thinkingmoonshotai/kimi-k2-thinkingazure_ai/kimi-k2-thinkingbedrock/moonshotai.kimi-k2-thinkinggmi/moonshotai/Kimi-K2-Thinkingcrusoe/moonshotai/Kimi-K2-Thinking
Frequently asked questions
Kimi K2 Thinking pricing, context and availabilityIs Kimi K2 Thinking still available?
Yes. Kimi K2 Thinking is scheduled for retirement on 2027-03-30.
How much does Kimi K2 Thinking cost?
Kimi K2 Thinking costs $0.60 per 1M input tokens from Moonshot, and $1.20 per 1M output tokens from Gmi. Across 10 serving providers, input prices range from $0.47 to $2.50 per 1M tokens.
What is the context window of Kimi K2 Thinking?
Kimi K2 Thinking accepts up to 262K tokens of context and can return up to 262K output tokens.
Which providers serve Kimi K2 Thinking?
Kimi K2 Thinking is available from 10 serving providers, each with its own pricing and model ID. Moonshot is currently the cheapest.
What can Kimi K2 Thinking do?
Kimi K2 Thinking supports tool use, reasoning, prompt caching, structured output and web search.
Other Moonshot models
compare pricing across the Moonshot lineup Listed prices from LiteLLM, OpenRouter and Vercel · last checked Sep 22, 2026
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI