Flow - 4x faster than typing
Mini tools| Model | Input | Cached input | Output | Unit |
|---|---|---|---|---|
|
Kimi K3
This model
Moonshot AI
|
$3 | $0.3 | $15 | per 1M tokens |
Kimi K3
By Moonshot AI
Kimi K3 is a sparse Mixture-of-Experts model with 2.8 trillion total parameters, activating 16 of 896 experts per token. It introduces two architectural changes: Kimi Delta Attention, a hybrid linear attention mechanism that speeds up decoding at million-token context lengths, and Attention Residuals, which selectively retrieves representations across network depth. It supports a context window of up to 1,048,576 tokens and accepts both text and image input through a single multimodal architecture. It runs with an always-on reasoning mode called thinking mode, currently fixed at maximum reasoning effort. K3 is built for long-horizon coding, research and document work, and agentic tool use, including operating tools reliably across extended multi-step tasks. It is served through an OpenAI SDK compatible API.
New
Multimodal
Overview
Kimi K3 is a 2.8 trillion parameter sparse Mixture-of-Experts language model with native vision understanding and a context window of up to 1 million tokens. It uses Kimi Delta Attention and Attention Residuals to improve long-context efficiency and depth scaling, with an always-on reasoning mode for coding, knowledge work, and agentic tool use.
Pricing
Compare Kimi K3 with other models listed in the same vendor pricing tiers and context lengths.
Standard
About Moonshot AI
Moonshot AI is a Beijing-based Chinese AI startup (founded 2023) focused on developing large language models and multimodal systems
View Company Profile
Last updated: July 17, 2026
