
Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at navigating large repositories, using tools, debugging, and iterating against images, logs, tests, and runtime feedback. Its architecture uses KDA and Attention Residuals for computational efficiency.
Modalities
In / Out Price
$2.60 / $13per 1M
Context
1M
Released
Jul 16, 2026
Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at navigating large repositories, using tools, debugging, and iterating against images, logs, tests, and runtime feedback. Its architecture uses KDA and Attention Residuals for computational efficiency.
Kimi K3 costs $2.60/M input tokens and $13.00/M output tokens, with separate rates for Cache Read at $0.29/M tokens.
Kimi K3 has a 1,048,576 token context window. It supports up to 974,842 completion tokens.
Yes. Kimi K3 accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.
Kimi K3 accepts text, images and video as input and returns text.
Kimi K3 is served by 12 providers on OpenRouter: Sail Research, Morph Fast, DeepInfra, DigitalOcean, Baseten, Together, Moonshot AI, Wafer and 4 more. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.
Kimi K3 was released on July 16, 2026.