API
Agents/Claude 4.5 Sonnet

Claude 4.5 Sonnet

by Anthropic

best score
71.4%
Best Score
bash-only
Best Leaderboard
1
Models Used
Yes
Open Source
EntryScore
Claude 4.5 Opus (high reasoning)76.8%
Claude Opus 4.675.6%
Claude 4.5 Sonnet (high reasoning)71.4%
Claude 4.5 Haiku (high reasoning)66.6%
mini-SWE-agent + Claude 4.5 Sonnet (high reasoning)71.4%
Claude 4.6 Opus72.0%
Claude 4.5 Opus70.7%
Claude 4.5 Sonnet67.0%
Claude 4.5 Haiku64.7%
Claude 4.5 Opus medium (20251101)74.4%
Sonar Foundation Agent + Claude 4.5 Sonnet74.8%
Claude 4.5 Sonnet (20250929)70.6%
mini-SWE-agent + Claude 4.5 Sonnet (20250929)70.6%
Claude 4 Opus (20250514)67.6%
Claude 4 Sonnet (20250514)64.9%
Claude 3.7 Sonnet (20250219)52.8%
Tools + Claude 3.7 Sonnet (2025-02-24)63.2%
Tools + Claude 3.5 Sonnet (2024-10-22)49.0%
Tools + Claude 3.5 Haiku40.6%