Kimi K3.1 Pro: benchmark results

Provider: Moonshot. Released 2026-10-23. Access: API.

Aggregate Rating 1934 ± 15, rank #12 of 1654 rated models, from 113 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA GDP.pdf - Legal - Criterion Pass Rate89.44Criterion Pass Rate (%)100
AA Terminal-Bench 4.067.17Pass@1 (%)100
Kimi K3.1 Pro Tech Blog - Agents' Last Exam (ALL)43.2Score (%)100
Kimi K3.1 Pro Tech Blog - Agents' Last Exam (CLI)44.2Score (%)100
Kimi K3.1 Pro Tech Blog - BenchCAD w/ tools92.7Voxel IoU (%)100
Kimi K3.1 Pro Tech Blog - BenchCAD w/o tools72.8Voxel IoU (%)100
Kimi K3.1 Pro Tech Blog - BioMysteryBench80.6Score (%)100
Kimi K3.1 Pro Tech Blog - BrowseComp92.1Score (%)100
Kimi K3.1 Pro Tech Blog - DECK-Arena1456Elo100
Kimi K3.1 Pro Tech Blog - GDP.pdf34.6Strict response pass rate (%)100
Kimi K3.1 Pro Tech Blog - GDP.pdf w/ python85.7Task-macro rubric pass rate (%)100
Kimi K3.1 Pro Tech Blog - HANDBOOK.md44.6Score (%)100

Interactive version: theaggregate.ai/model?slug=kimi-k3-1-pro · How It Works · Data refreshed daily, snapshot 2026-10-24.