Back to Models

MoonshotAI: Kimi K2.7 Code

Moonshotaitext
8.8 / 10 Overall Rating

MoonshotAI Kimi K2.7 Code is a long-context, coding-optimized Mixture-of-Experts (MoE) model with native vision support. It features a 262K token context window and an unusually high 235K token output limit designed for multi-file codebase generation. It operates at roughly a quarter of the API cost of GPT-4o and Claude 3.5 Sonnet.

Context Window
262K
Knowledge Cutoff
2025
Max Output
16K
License
Proprietary

Where it Excels (Pros)

  • 262K context window beats baselines
  • Massive 235K max token output
  • Significantly cheaper than GPT-4o
  • Multimodal vision input for coding

Limitations (Cons)

  • Niche focus on software development
  • Unproven general reasoning vs Sonnet

Benchmark Breakdown

Reasoning (MMLU)82%
Coding (HumanEval)78%
Mathematics (MATH)76%
Long-Context Retrieval95%

Pricing Matrix

Input cost / 1M tokens$0.66
Output cost / 1M tokens$3.40

Compare MoonshotAI: Kimi K2.7 Code with other frontier models:

Related Skills, Tools & Automations