Kimi K3
NewPremiumMoonshotAI's 2.8T-parameter flagship for complex coding, knowledge work, and long-horizon agent runs.
1M-token context with always-on reasoning and native image input. Built for navigating large repositories, using tools, debugging, and iterating against images, logs, tests, and runtime feedback. A Stable LatentMoE architecture with KDA and Attention Residuals keeps inference efficient at frontier scale.
▼
Kimi K3 is very affordable at $3.00/M tokens input and $15.00/M tokens output. Cached input reads are priced at $0.30/M tokens, making repeated context cheaper. The model supports vision/image input, extended reasoning, a 1M context window.
Benchmarks
Category Rankings
Specifications
Performance
Providers
Misc
Compare with
GPT 5.6 Solvs Kimi K3GPT 5.6 Lunavs Kimi K3
Claude Fable 5vs Kimi K3Claude Opus 4.8vs Kimi K3Claude Sonnet 5vs Kimi K3Claude Haiku 4.5vs Kimi K3
Gemini 3.1 Provs Kimi K3Gemini 3.5 Flashvs Kimi K3Gemini 3.1 Flash Litevs Kimi K3Gemma 4 31Bvs Kimi K3
Grok 4.5vs Kimi K3Grok Build 0.1vs Kimi K3Grok 4.1 Fastvs Kimi K3
Muse Spark 1.1vs Kimi K3
DeepSeek V4 Provs Kimi K3DeepSeek V4 Flashvs Kimi K3
MiniMax M3vs Kimi K3
GLM 5.2vs Kimi K3
Qwen3.7 Maxvs Kimi K3Qwen3.7 Plusvs Kimi K3Qwen3 Coder Nextvs Kimi K3
Nemotron 3 Ultravs Kimi K3
Mistral Small 4vs Kimi K3Frequently Asked Questions about Kimi K3▼
When was Kimi K3 released?
▼
Kimi K3 was released on July 16, 2026.
Who created Kimi K3?
▼
Kimi K3 was created by MoonshotAI.
How much does Kimi K3 cost?
▼
Kimi K3 costs $3.00/M input tokens and $15.00/M output tokens.
What is Kimi K3 API pricing?
▼
The Kimi K3 API is priced at $3.00 per million input tokens and $15.00 per million output tokens. You can access Kimi K3 through idapt.app alongside 200+ other AI models in one workspace.
Is Kimi K3 a reasoning model?
▼
Yes, Kimi K3 is a reasoning model. It can think step-by-step through complex problems before providing an answer, often yielding better results on difficult tasks like math, coding, and multi-step logic.
Does Kimi K3 support image or vision input?
▼
Yes, Kimi K3 supports vision/image input. You can share images in your conversation and the model will analyze and respond based on their content.
Does Kimi K3 support audio?
▼
No, Kimi K3 does not support audio input.
What is the context window of Kimi K3?
▼
Kimi K3 has a context window of 1M tokens, meaning it can process approximately 786,432 words of context at once.
Is Kimi K3 open source?
▼
Kimi K3 is a proprietary model developed by MoonshotAI. The model weights and training data are not publicly available.
What is Kimi K3 good at?
▼
Based on benchmark data, Kimi K3 excels at: visual analysis and image understanding. It is MoonshotAI's capable model suitable for a wide range of tasks.
Where can I use Kimi K3?
▼
You can chat with Kimi K3 on idapt.app — no separate API key required. idapt is an AI workspace with 200+ models, agents, computers, and tasks. You can also access Kimi K3 through MoonshotAI's own API or platform.