Model pricing
Kimi K2 Thinking pricing
Kimi K2 Thinking by Moonshot costs
Kimi K2 Thinking pricing
.2 per 1M input tokens and $5 per 1M output tokens (cached input $0.12, batch input not recorded) as of 2026-06-10. A 1,000-input/500-output request costs $0.0037. The lowest-cost alternative with at least this context window and the same recorded reasoning category is Gemini 4.5 Nano at $0.04/$0.16; capability equivalence is not guaranteed.Notes
Reasoning-tuned K2 variant — extended chain-of-thought for math, code, science
Key pricing facts
| Fact | Value |
|---|---|
| Input / 1M tokens | $1.2 |
| Output / 1M tokens | $5 |
| Cached input / 1M | $0.12 |
| Batch input / output / 1M | Not recorded |
| Context window | 262,144 |
| Provider | Moonshot |
| Last verified | 2026-06-10 |
| Snapshot updated | 2026-06-10 |
Last updated: · Source: Moonshot pricing · Methodology
Frequently asked questions
What does Kimi K2 Thinking cost per million tokens?
Input costs $1.2 and output costs $5 per million tokens. The 1,000-input/500-output example costs $0.0037, before discounts, taxes, hidden reasoning tokens or context-tier surcharges.
Does Kimi K2 Thinking have cached or batch pricing?
Cached input: $0.12 per million tokens under the existing provider cache rules. Batch: not recorded in this dataset. Eligibility and cache-write charges depend on the provider; these discounts are not automatically stacked.
What is a cheaper alternative to Kimi K2 Thinking?
Gemini 4.5 Nano costs $0.04 input and $0.16 output per million tokens. This comparison matches context capacity and reasoning category, not benchmark quality. Test your workload before replacing a model.
How often are these prices verified?
The live pricing table synchronizes stored model records daily; public file snapshots update when the site is rebuilt. Synchronization is not independent provider verification. Snapshot updated 2026-06-10. The recorded model verification date is 2026-06-10.