Intelligence index
Outperforms 92% of compared models
Kimi K3
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.
- Input / output price
- $0.159/$11.5per 1M tokens
- Context
- 1.0M
- Released
- 2026-07-16
Pricing
Artificial Analysis
Explore rankingsCoding index
Outperforms 93% of compared models
Agentic index
Outperforms 88% of compared models
Design Arena
Elo scores and win rates across design and agent tasks.
Category performance
1,000-1,500 EloModel Arena
Code Categories
1359Elo
Win rate 62.2%
Website
1335Elo
Win rate 59.3%
Data Visualization
1351Elo
Win rate 63.2%
SVG
1302Elo
Win rate 59.7%
3D
1385Elo
Win rate 67.2%
Game Development
1362Elo
Win rate 60.3%
UI Component
1334Elo
Win rate 59.5%
Agent Arena
Fullstack
1315Elo
Win rate 61.8%
Agenticgamedev
1216Elo
Win rate 54.9%
Androidnative
1253Elo
Win rate 56.3%
Mobile Apps
1267Elo
Win rate 57.6%
Htmlslides
1153Elo
Win rate 55.6%
Blender3d
1215Elo
Win rate 50.3%
Python-pptxslides
1219Elo
Win rate 54.8%
Webapps
1295Elo
Win rate 59.3%
Spreadsheet
1114Elo
Win rate 32.4%
Godotgamedev
1223Elo
Win rate 48.2%
Related Models
Kimi K3.1 Pro
moonshotaiKimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D creation, financial research, and end-to-end agentic knowledge work, with native image and video understanding.
- Context
- 1.0M
- Max output
- 943.7K
- Released
- 2026-10-23
- Input
- $2/M
- Output
- $10/M
- Cache read
- $0.2/M
Kimi K2.6
moonshotaiKimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration. It handles complex end-to-end coding tasks across Python, Rust, and Go, and can convert prompts and visual inputs into production-ready interfaces. Its agent swarm architecture scales to hundreds of parallel sub-agents for autonomous task decomposition - delivering documents, websites, and spreadsheets in a single run without human oversight.
- Context
- 262K
- Max output
- 235.9K
- Released
- 2026-04-20
- Input
- $0.65/M
- Output
- $3.41/M
- Cache read
- $0.15/M
Kimi K2.7 Code
moonshotaiMoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts. It uses a native multimodal mixture-of-experts architecture that accepts text and image input, and it always operates in a thinking mode, preserving full reasoning content across multi-turn conversations. With a 256K-token context window, it targets long-horizon coding, agentic task decomposition, and multi-turn dialogue. The model activates 32B parameters out of roughly 1T total.
- Context
- 262K
- Max output
- 235.9K
- Released
- 2026-06-12
- Input
- $0.56/M
- Output
- $3.1/M
- Cache read
- $0.15/M