Moonshot AIopen weightsreasoningKimi K2.7 Code pricing
$0.67 per 1M input tokens and $3.35 per 1M output tokens. In real work that is 0.56¢ per chat answer and 1.3¢ to summarize a 20-page PDF.
Official price at Moonshot (Kimi): $0.95 / $4.00 per 1M, cached $0.19 (source, verified Oct 18, 2026). OpenRouter lists $0.67 / $3.35, which is the cheapest route across its hosts - lower than the official price.
What Kimi K2.7 Code costs per task
Token counts estimated from a measured model of the same lab (Gemini 3.8 Flash), rescaled by tokenizer. Reasoning tokens are billed as output.
| Task | Cost | Tokens used | Price rank | Basis |
|---|---|---|---|---|
| Answer a chat question | 0.56¢$5.57 / 1,000 | 36 in · 811 out · 843 reasoning | 232 of 336 | estimated |
| Reply to a customer email | 0.20¢$2.05 / 1,000 | 279 in · 131 out · 424 reasoning | 232 of 336 | estimated |
| Write a 1,500-word blog post | 1.2¢$12.30 / 1,000 | 53 in · 2,698 out · 964 reasoning | 200 of 336 | estimated |
| Write 10 product descriptions | 1.0¢$10.18 / 1,000 | 301 in · 1,160 out · 1,817 reasoning | 211 of 336 | estimated |
| Translate 1,000 words | 1.6¢$15.99 / 1,000 | 1,346 in · 1,335 out · 3,167 reasoning | 221 of 336 | estimated |
| Proofread one page (500 words) | 0.42¢$4.22 / 1,000 | 726 in · 672 out · 443 reasoning | 217 of 336 | estimated |
| Summarize a 20-page PDF | 1.3¢$13.04 / 1,000 | 13,117 in · 550 out · 713 reasoning | 211 of 336 | estimated |
| Extract data from an invoice | 0.53¢$5.33 / 1,000 | 465 in · 554 out · 943 reasoning | 231 of 336 | estimated |
| Analyze a 200-row spreadsheet | 3.2¢$31.65 / 1,000 | 6,490 in · 1,048 out · 7,100 reasoning | 193 of 336 | estimated |
| Answer from 5 retrieved documents (RAG) | 0.98¢$9.82 / 1,000 | 3,714 in · 688 out · 1,499 reasoning | 231 of 336 | estimated |
| Classify 100 support tickets | 2.3¢$23.23 / 1,000 | 1,438 in · 587 out · 6,059 reasoning | 273 of 336 | estimated |
| Review 500 lines of code | 1.1¢$10.75 / 1,000 | 5,360 in · 2,135 out | 157 of 336 | estimated |
| Write a SQL query | 0.55¢$5.47 / 1,000 | 138 in · 562 out · 1,043 reasoning | 234 of 336 | estimated |
| Run a 50-step coding agent | 66¢$656 / 1,000 | 2,259,875 in · 15,150 out · 47,150 reasoning | 193 of 336 | modeled |
Price at every provider
List prices per 1M tokens (input / output) at each host serving Kimi K2.7 Code through OpenRouter today. Hosts can differ in quantization, context and speed.
| Provider | Input / output | Cached input | Context | Uptime |
|---|---|---|---|---|
| S StreamLakeunknown | $0.71 / $3.00 | $0.14 | 256K | 100% |
| I Inceptronint4 | $0.67 / $3.35 | $0.18 | 262K | 100% |
| C CoreWeaveint4 | $0.71 / $3.50 | $0.15 | 262K | 100% |
| V Veniceint4 | $0.75 / $3.50 | $0.16 | 256K | 100% |
 SiliconFlowfp8 | $0.86 / $3.80 | $0.18 | 262K | 100% |
 Novitaint4 | $0.91 / $3.84 | $0.18 | 262K | 100% |
| P Phalaunknown | $0.95 / $4.00 | $0.19 | 262K | 100% |
 Nebiusfp4 | $0.95 / $4.00 | - | 262K | 100% |
| G GMICloudfp8 | $0.95 / $4.00 | $0.19 | 262K | |
 Alibabafp8 | $0.95 / $4.00 | $0.19 | 262K | |
| C Cloudflareunknown | $0.95 / $4.00 | $0.19 | 262K | 100% |
Moonshot AIint4 | $0.95 / $4.00 | $0.19 | 262K | 100% |
Moonshot AIint4 | $1.90 / $8.00 | $0.38 | 262K | 100% |
Price history
First seen in our price history on Jun 16, 2026 at $0.75 input / $3.50 output. Since then the output price moved from $3.50 to $3.35 per 1M tokens (10 changes).
FAQ
How much does Kimi K2.7 Code cost?
Kimi K2.7 Code costs $0.67 per 1 million input tokens and $3.35 per 1 million output tokens, with cached input at $0.18 (list price on Oct 18, 2026). In practice one chat answer costs about 0.56¢ and summarizing a 20-page PDF about 1.3¢.
Where is Kimi K2.7 Code cheapest?
Among 13 providers we track, the lowest list price today is at StreamLake: $0.71 input and $3.00 output per 1M tokens.
Does Kimi K2.7 Code charge for reasoning tokens?
Yes. Reasoning (thinking) tokens are billed as output tokens. That is why the cost per task can be much higher than the visible answer suggests - our task costs include them.
Is the Kimi K2.7 Code cost per task measured or estimated?
Estimated: Kimi K2.7 Code was not run directly. Tokens come from Gemini 3.8 Flash and are rescaled by the tokenizer of Moonshot AI. Real usage can differ.