Kimi K2 Thinking
Moonshot AI · model · context window 262,144 tokens.
Pricing (per 1M tokens)
Specifications
Intelligence & value
MeasuredWorth it? Qwen3.5 Flash matches or beats Kimi K2 Thinking on the overall Arena and costs about $18.70/month versus $215.00/month at 100,000 calls — roughly $196.30/month cheaper.
Intelligence scores are community Arena ratings from LMArena / arena.ai, used under CC BY 4.0. Snapshot last refreshed 18 September 2026. Scores are a relative signal, not an absolute measure of capability. A Measured score was voted on directly; an Estimated score is inherited from an identical base model (a regional/creator-prefixed hosting duplicate). See our methodology.
Where else you can run Kimi K2 Thinking
The same model is resold by 17 providers we track. The typical one charges 1.3× what the cheapest does — $1.08 against $0.8530 per 1M tokens, blended 3:1. Routing the same workload to a different host is often a bigger saving than switching model.
| Provider | Input / 1M | Output / 1M | Blended |
|---|---|---|---|
| Vercel AI Gateway | $0.4700 | $2.00 | $0.8530 |
| Helicone | $0.4800 | $2.00 | $0.8600 |
| Opencode | $0.4000 | $2.50 | $0.9250 |
| IO NET | $0.5500 | $2.25 | $0.9750 |
| Alibaba CN | $0.5740 | $2.29 | $1.00 |
| 302.AI | $0.5750 | $2.30 | $1.01 |
| Huggingface | $0.6000 | $2.50 | $1.08 |
| Hyper | $0.6000 | $2.50 | $1.08 |
| Kilo | $0.6000 | $2.50 | $1.08 |
| LLM Gateway | $0.6000 | $2.50 | $1.08 |
Showing the 10 cheapest of 17. The full list for every model is in the open endpoint dataset.
Third-party listings via models.dev, not rates we verify with each host the way we do first-party pricing. Treat them as a shortlist to check, not a quote — and confirm quantisation, context limits and rate limits before switching, since a cheaper host is not always serving the same thing.
Estimate cost with Kimi K2 Thinking
Kimi K2 Thinking is pre-selected below. Enter your token counts and volume to see the cost per request, per day and per month — then switch models to compare.
Results update automatically as you type.
- Input cost
- $0.000600
- Output cost
- $0.00125
- Per request
- $0.00185
- Per day
- $1.85
- Per month
- $55.50
Source: auto-discovery (models.dev). Prices are re-checked daily against the provider’s official pricing; this rate was last changed on 19 July 2026. A recent date here just means the price is still current, not that it moved.