Context
1M
Max output
1M
Serving providers
14
Cheapest input
$0.70 /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. An OpenRouter routed quote is the price of whichever upstream provider OpenRouter prefers today, so it moves day to day.
Benchmarks
independent evaluations · source: Artificial AnalysisSpeed & latency
Output speed
45 t/s
tokens / second
Time to first token
3.17s
latency
Headline indices
Individual benchmarks · click a row for effort variants
| Benchmark | Best score | Effort | Cost / task | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Aa Analyst Agent | 0.4 | default | — | |||||||||||||||||
| APEX Agents | 41% | default | — | |||||||||||||||||
| Automation Bench | 0.6 | default | — | |||||||||||||||||
| CritPt | 23% | default | — | |||||||||||||||||
| EvalComputeProxyFrozen Aug 18, 2026 | 13706.1 | default | — | |||||||||||||||||
| GDPval v2Frozen Aug 18, 2026 | 1681.1 | default | — | |||||||||||||||||
| ||||||||||||||||||||
| GPQA Diamond | 94% | default | — | |||||||||||||||||
| Harvey Lab | 0.1 | default | — | |||||||||||||||||
| Humanity's Last Exam | 47% | default | — | |||||||||||||||||
| IT-Bench SRE | 48% | default | — | |||||||||||||||||
| Long-Context Reasoning | 89% | default | — | |||||||||||||||||
| Mlcr OverallFrozen Sep 9, 2026 | 0.4 | default | — | |||||||||||||||||
| MMMU-Pro | 81% | default | — | |||||||||||||||||
| Omniscience | 19.7 | default | — | |||||||||||||||||
| Omniscience Accuracy | 0.5 | default | — | |||||||||||||||||
| Omniscience Non Hallucination | 0.5 | default | — | |||||||||||||||||
| SciCode | 59% | default | — | |||||||||||||||||
| Terminal Bench40 | 0.1 | default | — | |||||||||||||||||
| ||||||||||||||||||||
| TerminalBench v2.1 | 85% | default | — | |||||||||||||||||
| ||||||||||||||||||||
| τ-bench Banking | 46% | default | — | |||||||||||||||||
Artificial Analysis last published scores for this benchmark on Sep 9, 2026. Scores here are frozen at that date, and models released since are not on it.
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 $0.70 to $3.00 per 1M tokens across 14 providers, so the dearest route costs 329% more than the cheapest for the same model.
| Serving provider | Input /1M | Output /1M | Endpoints | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OpenRouterRouted quote | $0.70 | $15.00 | 1 | ||||||||||||||||
Standardmoonshotai/kimi-k3
| |||||||||||||||||||
| Sail | $2.50 | $12.50 | 1 | ||||||||||||||||
Standardsail/moonshotai/Kimi-K3
| |||||||||||||||||||
| DeepInfra | $2.85 | $14.25 | 1 | ||||||||||||||||
Standarddeepinfra/moonshotai/Kimi-K3
| |||||||||||||||||||
| Databricks | $3.00 | $15.00 | 1 | ||||||||||||||||
Standarddatabricks/databricks-kimi-k3
| |||||||||||||||||||
| Aihubmix | $3.00 | $15.00 | 1 | ||||||||||||||||
Standardaihubmix/kimi-k3
| |||||||||||||||||||
| Baseten | $3.00 | $15.00 | 1 | ||||||||||||||||
Standardbaseten/moonshotai/Kimi-K3
| |||||||||||||||||||
| Amazon Bedrock | $3.00 | $15.00 | 1 | ||||||||||||||||
Standardglobal.moonshotai.kimi-k3
| |||||||||||||||||||
| Fireworks AI | $3.00 | $15.00 | 1 | ||||||||||||||||
Standardfireworks_ai/accounts/fireworks/models/kimi-k3
| |||||||||||||||||||
| Vercel is dearest at $4.50 / $22.50 | |||||||||||||||||||
Price history
listed seller prices, input and output per 1M tokens Input Output
Input down 17% since first tracked
Cost calculator
estimate your monthly spend on this model$1,340
estimated / month
Model IDs
copy the exact identifier for your platformmoonshotai/kimi-k3moonshotai/kimi-k3:batchsail/moonshotai/Kimi-K3deepinfra/moonshotai/Kimi-K3databricks/databricks-kimi-k3tencent/kimi-k3together_ai/moonshotai/Kimi-K3moonshot/kimi-k3fireworks_ai/accounts/fireworks/models/kimi-k3novita/moonshotai/kimi-k3nebius/moonshotai/Kimi-K3aihubmix/kimi-k3moonshotai/kimi-k3global.moonshotai.kimi-k3baseten/moonshotai/Kimi-K3moonshotai/kimi-k3-fast
Frequently asked questions
Kimi K3 pricing, context and availabilityHow much does Kimi K3 cost?
Kimi K3 costs $0.70 per 1M input tokens as an OpenRouter routed quote, while Moonshot lists it at $3.00 per 1M input, and $12.50 per 1M output tokens from Sail. An OpenRouter routed quote is the price of whichever upstream provider OpenRouter prefers today, so it moves day to day. Across 14 serving providers, input prices range from $0.70 to $3.00 per 1M tokens.
What is the context window of Kimi K3?
Kimi K3 accepts up to 1M tokens of context and can return up to 1M output tokens.
Which providers serve Kimi K3?
Kimi K3 is available from 14 serving providers, each with its own pricing and model ID. OpenRouter is currently the cheapest, as a routed quote.
What can Kimi K3 do?
Kimi K3 supports image input (vision), tool use, reasoning, prompt caching and structured output.
Other Moonshot models
compare pricing across the Moonshot lineupKimi K2.7 Code$0.671/1M · 16 providersKimi K2.6$0.65/1M · 15 providersKimi K2.5$0.45/1M · 11 providersKimi K2 Thinking$0.60/1M · 10 providersKimi K2 Instruct$0.50/1M · 6 providersKimi K2 Instruct 0905$0.50/1M · 5 providersKimi K2.7 Code Highspeed$1.90/1M · 3 providersKimi Latest$0.56/1M · 2 providers
Popular comparisons
head-to-head pages featuring Kimi K3
Kimi K3 vs Claude Fable 5.1side-by-side pricing & benchmarks Kimi K3 vs Claude Opus 5side-by-side pricing & benchmarks
Kimi K3 vs DeepSeek V4.1 Flashside-by-side pricing & benchmarks Kimi K3 vs DeepSeek V4 Pro 0423side-by-side pricing & benchmarks
Kimi K3 vs Gemini 3.7 Flashside-by-side pricing & benchmarks Listed prices from LiteLLM, OpenRouter and Vercel · last checked Oct 14, 2026
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI