moonshotai/kimi-k2 vs moonshotai/kimi-k2-thinking
For 100,000 calls of 2,000 in and 500 out, moonshotai/kimi-k2 is cheaper: $229.00 a month at OpenRouter, against $245.00 for moonshotai/kimi-k2-thinking at Novita AI — 1.1× the cost.
Cheapest on each
moonshotai/kimi-k2
$229.00/mo
cheapest, at OpenRouter — $0.57 in, $2.3 out per million
moonshotai/kimi-k2-thinking
$245.00/mo
cheapest, at Novita AI — $0.6 in, $2.5 out per million
Aligned facts
Same rows on both sides, including the ones we cannot fill.
| Attribute | moonshotai/kimi-k2 | moonshotai/kimi-k2-thinking |
|---|---|---|
| Context window | 131K | 262K |
| Accepts | Aa | Aa |
| Sellers | 1 | 2 |
| Cheapest input rate | $0.57 / M | $0.6 / M |
| Cached input rate | ○ no answer | $0.15 / M |
No quality or benchmark row appears here. We hold no throughput or accuracy measurement for either model, and a cost-per-quality figure built from a lab’s own claim would look authoritative and mean nothing.
Cost for this workload — both models, every seller
100,000 calls a month · 2,000 input and 500 output tokens each. Sellers of both models lead, cheapest combined cost first.
| Seller | moonshotai/kimi-k2 /mo | moonshotai/kimi-k2-thinking /mo |
|---|---|---|
| OpenRouter | $229.002026-10-27 | $245.002026-10-27 |
| Novita AI | ○ no answer | $245.002026-10-22 |
Cheapest tier per seller, this workload only. In and output rates, and a cached prefix, are broken out in the cost calculator.