Sakura AI Engine gives you 3,000 free requests a month for Kimi-K2.6 and K2.7-Code
Sakura Internet's AI Engine has a free plan with 3,000 text generation requests per month, shared across Kimi-K2.6, Kimi-K2.7-Code, gpt-oss-120b and more, and the allowance resets every month.
Free plan with 3,000 chat requests a month, including Kimi-K2.6 and Kimi-K2.7-Code
- Eligible plans：
- Free foundation-model plan
- Eligibility：
- Phone verification to create a Sakura Internet member ID and a registered credit card for the Sakura Cloud project; free-plan sign-ups are capped
- Benefit duration：
- Allowance is per calendar month and resets automatically
- Renewal terms：
- Beyond the free allowance requests are rate-limited, not billed; the free plan never switches to pay-as-you-go automatically
The briefing
Sakura AI Engine is an inference API from Japanese cloud provider Sakura Internet. Models run in data centers inside Japan, and inputs are never stored or used for training. Its free foundation-model plan gives you 3,000 text generation requests a month at no cost, and the allowance resets every month.
What you can use
The 3,000 chat requests are shared across several models, currently including Kimi-K2.6, Kimi-K2.7-Code, gpt-oss-120b, Qwen3.6-35B-A3B, gemma-4-31B-it, Qwen3-VL-30B-A3B-Instruct and llm-jp-3.1. The Kimi, Qwen3 and Gemma models are public previews that can change or be withdrawn at any time. The medical model Weblab-MedLLM is not included in the free allowance.
The free plan also includes 50 speech-to-text requests (whisper-large-v3-turbo), 10,000 embedding requests (multilingual-e5-large) and 50 VOICEVOX text-to-speech requests per month.
The API is compatible with OpenAI Chat Completions and Responses as well as the Anthropic Messages API, so tools such as Claude Code can connect directly. The Messages endpoint currently supports Kimi-K2.6 only.
Once you use up the monthly allowance, requests are rate-limited rather than billed, and the free plan never switches to pay-as-you-go on its own. During busy periods you may get a temporary 429 error even with allowance left; wait a few minutes and retry. Sign-ups for the free plan are capped, and new applications stop once the cap is reached.
How to start
- Create a Sakura Internet member ID. Sign-up requires phone number verification.
- Log in to the control panel and create a Sakura Cloud project. This step requires registering a credit card.
- Open the AI Engine control panel, subscribe to the free foundation-model plan, issue an API token, then try models in the Playground or call them from any OpenAI- or Anthropic-compatible client.
See the Sakura AI Engine page for details.