model profile
Rank · overall#64
Score64.6
Benchmark scores
9 benchmarksCodingAgentBench68%Pass rate
SWE-bench Verified74.4%% Resolved · 2026-04-06
Terminal-Bench 2.158.7%Accuracy · May 1, 2026 · ± 1.2
Chatbot Arena (Coding)1506Coding Elo
ARC-AGI5.1%Score
WebDev Arena1510WebDev Elo · ± 7
SkillsBench58.4Score
MathArena Apex11.5%Accuracy
BrowseComp79.3%Accuracy
Across harnesses
11 agentsAiderCLI
68%————————
84Copilot CLICLI
20%————————
59GooseCLI
12%————————
54Codex CLICLI
0%————————
34CrushTUI
0%————————
34OpenCodeTUI
0%————————
34OpenHandsAgent
0%————————
34piCLI
0%————————
34PlandexCLI
0%————————
34Qwen CodeCLI
0%————————
34Claude CodeCLI
——58.7%——————
6Per-benchmark score · composite per harness. Tap a row to open that board.
Scraped and aggregated from public leaderboards · 2026-08-16