Moonshot AI
Moonshot AI’s premier flagship model with 2.8 trillion parameters, Kimi Delta Attention, 1M context, and open weights.
1.049M ctx$3.00 / 1M tok (2.8T Open Weights)
Introducing Kimi K2.7 CodeMoonshot AIReleased June 10, 2026
Kimi K2.7 Code
Dedicated coding model delivering higher task success rates, tighter instruction following, and a 30% reduction in overthinking.
reasoning-llmProprietary API$0.95 / 1M tok (Dedicated Coding)Context: 262.144K (262,144 tokens)Arena ELO: 1485
Technical Specifications
Architecture Type
Dedicated Autonomous Coding Transformer
Total Parameters
Frontier Scale
Context Window
262.144K (262,144 tokens)
Max Output Tokens
65.536K (65,536 tokens)
Knowledge Cutoff
May 2026
Supported Modalities
text, code, vision, reasoning
License & Access
Moonshot AI Terms of Service
Benchmark Evaluations
Mmlu Pro
88
Humaneval
97.4
Swe Bench Verified
76.5
Chatbot Arena ELO
1485
Deep Architectural Overview
Kimi K2.7 Code is tailored specifically for software engineers and autonomous programming agents (Claude Code, OpenCode, Cline). It drastically curbs circular reasoning and overthinking loops while reliably adhering to project conventions across 256k tokens.
Strengths & Considerations
Core Strengths
- Dedicated coding optimization with 30% less overthinking
- 76.5% SWE-bench Verified coding score
- 256k context window and 64k output buffer
- Affordable $0.95 input / $0.19 cache hit pricing
Known Limitations
- Optimized primarily for code and STEM; general creative writing better on K3
Token & API Pricing
Input Tokens (1M)$0.95
Output Tokens (1M)$4.00
Cached Input (1M)$0.1900
Pricing is verified directly against Moonshot AI's developer documentation and API rate sheets.
Similar & Alternative Models
Explore other frontier models from Moonshot AI and comparable reasoning engines.
Moonshot AI
High-speed variant of Kimi K2.7 Code with generation speeds of 180 to 260 tokens per second for instantaneous coding feedback.
262.144K ctx$1.90 / 1M tok (180-260 tok/s)
Moonshot AI
General-purpose model supporting text, image, and video inputs, switchable thinking modes, and autonomous agent tasks over 256k context.
262.144K ctx$0.95 / 1M tok (Vision & Agentic)