Moonshot AI · LIVEBENCH 2026-06-25

Kimi K2.7 Code

Published benchmark results for this specific model configuration.

kimi-k2.7-code

Overall score68.4 / 100

Cost / successful task$0.100USD · benchmark workload

Weight accessOpen weights

Where this configuration performs

Category averages from the same release, on a 0–100 scale.

Reasoning82.8

Coding74.0

Agentic Coding45.7

Mathematics79.6

Data Analysis62.7

Language77.9

Instruction Following56.3

Task-level results

theory of mind69.2zebra puzzle96.0spatial92.0logic with navigation74.0code generation71.8code completion76.1javascript63.6typescript23.3python50.0AMPS Hard97.0integrals with game37.0math comp94.1olympiad90.3consecutive events48.3tablejoin47.5tablereformat92.2connections98.0plot unscrambling55.7typos80.0paraphrase48.5simplify49.8story generation68.6summarize58.3

Source files: Scores ↗ Categories ↗ Costs ↗ . Missing cost is unknown, never free. Compare configurations within the same release.