Moonshot AI
Kimi K2 API
The Kimi API docs list Kimi K2.6 as the current agentic model with text, image and video input, 256K context and thinking/non-thinking modes, alongside K2.5, K2 and Moonshot V1 models. The platform documents OpenAI API compatibility, streaming, multimodal input and pay-as-you-go pricing.
Quick answers
At a glance
- Overview
- Kimi's international developer API for K2.6, K2.5, K2 and Moonshot V1 models with OpenAI-compatible access.
- Best fit
- Developers evaluating Chinese long-context, coding and agentic models through an international API.
- Trust
- 3/3 sources verified, recently checked · 2026-05-17
- Coverage
- 100/100
Editorial verdict
Best for
Developers evaluating Chinese long-context, coding and agentic models through an international API.
Avoid if
Avoid assuming consumer Kimi behavior or limits match API behavior.
Why it matters
Kimi's developer path is now separated through K2.6 and the English API platform, so it should be distinct from the consumer assistant.
Pricing
Pay-as-you-go API pricing varies by Kimi K2 and Moonshot model
Payment
Pay-as-you-go API billing, Platform billing, Enterprise sales
Commercial use
Commercial use should follow the current product, API, model license and billing terms.
Privacy
Review prompt, file, media upload, retention and training-use terms before sensitive workloads.
Use-case fit
Long-context API evaluation
StrongUse K2.6 or Moonshot V1 for long-context prompts, file-like workloads and complex reasoning checks.
Multimodal agent API
StrongK2.6 is documented with text, image and video input for agentic tasks.
Global user checklist
Model names, quotas, release status, regional access and commercial terms can change quickly; recheck official sources before procurement or production use.
Pros
- - K2.6 supports 256K context and multimodal input
- - OpenAI-compatible API path is documented
- - Pricing and model lists are English-facing
Cons
- - Model generations and preview status can change quickly
- - Consumer Kimi and API platform limits are separate
Decision paths
qwen
deepseek
zhipu-glm