Models
DeepSeek, Kimi, Qwen and MiniMax API
Call DeepSeek V4, Kimi K3 and K2.7 Code, Qwen 3.8 Flash and MiniMax M3 through one OpenAI-compatible API key. Compare per-token prices in one table.
tokely.me/chinese-ai-models-api
Kimi API: Kimi K3 and K2.7 Code
Kimi K2.5: $0.825 input / $4.33125 output per 1M tokens. Up to 10% off selected models vs Moonshot prices. Pay as you go.
By TokelyReviewed Kimi K3 and Kimi K2.7 Code are available through Tokely's OpenAI-compatible API. K2.7 Code is tuned for coding agents; K3 is the general flagship. Both bill per token from your Tokely balance with cache-read pricing for repeated context.
Kimi API price comparison
| Model | Tokely input / output | Tokely example | model developer example | Discount vs model developer |
|---|---|---|---|---|
| Kimi K2.5Moonshot | $0.825 / $4.33125Cache read: $0.825 | $1.907813 | Not matched | — |
| Kimi K2.6Moonshot | $1.30625 / $5.426025Cache read: $0.221058 | $2.662756 | $1.95$0.95 / $4 per 1M | 36.6% higher |
| Kimi K2.7 CodeMoonshot | $1.30625 / $5.426025Cache read: $0.26125 | $2.662756 | $1.95$0.95 / $4 per 1M | 36.6% higher |
| Kimi K3Moonshot | $2.7 / $13.5Cache read: $0.27 | $6.075 | $6.75$3 / $15 per 1M | 10.0% off |
K2.7 Code for agents
Kimi K2.7 Code was tested through api.tokely.me with function tools, streaming usage, JSON mode, reasoning effort, inline images and prompt caching. Use it in coding tools that accept an OpenAI-compatible provider.
K3 for general work
Kimi K3 is the general-purpose flagship for long-context chat and analysis. Compare both in the table and with your own token mix in the calculator.
One key for every model
Every model on this page is available through one OpenAI-compatible endpoint and one balance, with routing modes, bounded fallback, spending limits and request logs.
LLM gateway: one API for every model ↗
A hosted LLM gateway with one OpenAI-compatible endpoint, account routing modes, bounded fallback, spending limits and request logs. Text, image, video and audio from one balance.
OpenAI-compatible API for Claude, Gemini, GPT and more ↗
Keep the OpenAI SDK and call Claude, Gemini, GPT, Kimi, GLM and other supported models. Change the base URL, key and model ID.
Questions before you switch
Is the Kimi API OpenAI-compatible?
Yes on Tokely: base URL https://api.tokely.me/v1 with model kimi-k3 or kimi-k2.7-code.
START WITH YOUR OWN WORKLOADKnow the price.
Know the price.
Then make the call.
Check the model, compare the rate, and connect with a Tokely key.