Moonshot AI › Kimi K2.7 Code › Pricing
Kimi K2.7 Code pricing
API list price by Moonshot AI · checked Oct 23, 2026
$0.95 / $4per 1M tokens in / out
$0.19cache read · −80%
$0.54per task at thinking?
#111on OpenRouter by tokens · 57.5B / week
Price sheetper 1M tokens
| Input | $0.95 | list |
| Output | $4 | |
| Cache read | $0.19 | −80% vs input |
| Context · max output | 256K · 236K | tokens |
Where to buy16 providers · per 1M tokens
| Provider | Input | Output | Context |
|---|---|---|---|
| Inceptron | $0.67 | $3.35 | 262K |
| DeepInfra | $0.68 | $3.40 | — |
| CoreWeave | $0.71 | $3.50 | 262K |
| Venice | $0.75 | $3.50 | 256K |
| Featherless | $0.80 | $3.40 | — |
| SiliconFlow | $0.86 | $3.80 | 262K |
| lab-qwen | $0.89 | $3.71 | — |
| Alibaba | $0.95 | $4 | 262K |
| Cloudflare | $0.95 | $4 | 262K |
| GMICloud | $0.95 | $4 | 262K |
| lab-moonshotai | $0.95 | $4 | — |
| Moonshot AIlab | $0.95 | $4 | 262K |
| moonshotai | $0.95 | $4 | — |
| Nebius | $0.95 | $4 | 262K |
| Novita | $0.95 | $4 | — |
| Phala | $0.95 | $4 | 262K |
In Moonshot AI's line-upper 1M in / out
| Kimi K2.7 Code | $0.95 / $4 | this model |
| Kimi K2.7 Code Highspeed | $1.90 / $8 | 2× |
| Kimi K3.1 Pro | $2 / $10 | 2.3× |
| Kimi K3 | $3 / $15 | 3.5× |
| Kimi K3.1 Pro Highspeed | $4 / $20 | 4.7× |
Cheaper for the same workvs Kimi K2.7 Code
| GPT-6.1 Sol xhigh | -27% | score +98% | $0.39 |
| MiMo v2.6 Pro thinkingOpen | -75% | score +79% | $0.13 |
| DeepSeek V4.1 Flash 1017 maxOpen | -52% | score +75% | $0.26 |
| Claude Haiku 5.5 max | -61% | score +68% | $0.21 |
| MiniMax M3.1 Flash thinking | -59% | score +65% | $0.22 |
Pricing questions
How much does Kimi K2.7 Code cost?›
$0.95 per 1M input tokens and $4 per 1M output tokens. Cached input is $0.19. One Intelligence Index task costs about $0.54 at the default effort.
Is Kimi K2.7 Code free to use?›
There is no free API tier, but the weights are open: you can run it on your own hardware for the cost of compute.
How much does prompt caching save?›
Cache reads cost $0.19 per 1M, 80% below the input price.
What is a cheaper alternative to Kimi K2.7 Code?›
GPT-6.1 Sol at xhigh effort costs -27% per task with a score +98% on Humanity's Last Exam.