| Provider | Input / 1M | Output / 1M | Monthly |
|---|
Calculating Kimi API Costs
Kimi models from Moonshot AI bill per token, with one twist that matters more than any other AI provider: the same model is sold at meaningfully different prices depending on where you buy it. Enter your request size and monthly volume above and the calculator prices your workload at Moonshot’s official rates, then shows what the identical usage costs on OpenRouter, DeepInfra, and Together below the result.
A worked example: a coding agent on Kimi K2.7 Code sending 100,000 input tokens and generating 10,000 output tokens per run costs $0.135 per run at official rates with no cache hits. At 10,000 runs a month that is $1,350 on Moonshot – or about $1,090 for the same runs through OpenRouter’s listed rate. The gap is the point of checking.
Kimi K2 Pricing
Official Moonshot international platform rates, checked July 3, 2026. All three current models take 262,144 tokens of context.
| Model | Input (cache hit) | Input (cache miss) | Output |
|---|---|---|---|
| kimi-k2.7-code | $0.19 | $0.95 | $4.00 |
| kimi-k2.7-code-highspeed | $0.38 | $1.90 | $8.00 |
| kimi-k2.6 | $0.16 | $0.95 | $4.00 |
| kimi-k2.5 | $0.10 | $0.60 | $3.00 |
Per 1 million tokens, USD, taxes excluded and calculated at checkout. HighSpeed is the same K2.7 Code model served faster at double the price.
K2.7 Code is Moonshot’s current flagship for coding and agent work; K2.6 is the general-purpose predecessor at the same list price, and K2.5 stays available as the budget option. Several comparison sites still present K2.6 as the newest model – the K2.7 Code pricing page has been live since late June 2026.
Moonshot also runs a separate Chinese platform (platform.kimi.com) with its own RMB pricing: K2.7 Code is ¥6.50 input / ¥27.00 output per million there. Billing accounts, currency, and tax treatment differ, so the two platforms are not interchangeable even though the models are.
Provider Price Differences
Because the K2 models are open-weight, third-party hosts run them and set their own prices.
Rates observed July 3, 2026, per million tokens:
| Provider | K2.7 Code in / out | K2.6 in / out | K2.5 in / out |
|---|---|---|---|
| Moonshot (official) | $0.95 / $4.00 | $0.95 / $4.00 | $0.60 / $3.00 |
| OpenRouter | $0.74 / $3.50 | $0.66 / $3.41 | $0.375 / $2.025 |
| DeepInfra | $0.74 / $3.50 | $0.75 / $3.50 | $0.45 / $2.25 |
| Together | $0.95 / $4.00 | $1.20 / $4.50 | not listed |
Two caveats before switching providers on price alone. OpenRouter routes requests across underlying hosts, so its listed price reflects current routing and can change without notice. And third-party hosts run the open-weight release, which can lag the official platform on model updates, context handling, or caching behavior – cheap tokens from a host with no cache support can cost more in practice than official tokens at an 80% hit rate.
Cache Hit Pricing
Moonshot caches repeated context automatically on its platform. Cached input on K2.7 Code bills at $0.19 per million instead of $0.95 – a 5x reduction, and the single biggest lever on a coding-agent bill, where the repeated project context usually dwarfs the new instructions in each request. Multi-turn agent sessions typically land high hit rates; independent one-shot requests land near zero. The pattern is the same one that drives DeepSeek’s cache-hit pricing, though DeepSeek’s discount is steeper at 50x.
Kimi Prices Compared With Other APIs
At official rates, K2.7 Code sits between the budget Chinese APIs and Western mid-tier models: roughly 7 times more expensive than DeepSeek V4 Flash ($0.14 / $0.28), about the same input price as Claude Haiku 4.5 but slightly cheaper on output ($4.00 vs $5.00), and roughly a third of Claude Sonnet’s standard rate ($3.00 / $15.00). For a Claude-side workload comparison with caching TTLs and batch discounts modeled properly, run the same numbers through the Claude API token cost calculator.
Keeping the Bill Down
Keep the stable part of your prompt byte-identical between requests so Moonshot’s cache catches it. Use K2.5 for classification and reformatting jobs – it is 37% cheaper on input and handles routine work; save K2.7 Code for the agentic coding it was built for. Skip HighSpeed unless latency genuinely costs you money: it doubles every token. And if you are provider-shopping, re-check prices at the source before committing – third-party rates in this market have moved monthly, and the table above is a July 3, 2026 snapshot, not a guarantee.
