aitoolcompass.

Find your next AI tool

Search by product name or task. Press Escape to close.

← All model benchmarks

Moonshot AI · LIVEBENCH 2026-11-06

kimi-k3.1-pro

Published benchmark results for this specific model configuration.

kimi-k3.1-pro
Overall score72.6 / 100
Cost / successful task$1.015USD · benchmark workload
Weight accessNot reported

Where this configuration performs

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

Long Agentic38.7
Reasoning91.1
Coding67.8
Browser75.9
Data Analysis79.1
Research82.1
Design73.7

Task-level results

long agentic38.7reasoning91.1coding67.8browser75.9data analysis79.1research82.1design73.7

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