List your company

Coding on Terminal-Bench 2.0

59.3ScoreClaude Opus 4.5
31.7438.89546.0553.205Feb 17, 2026

Evaluation results

MethodLinks
Claude Opus 4.5Agent Framework=Terminus-22026.0259.3—
Claude Opus 4.5Agent Framework=Claude Code2026.0257.9—
GLM-5Agent Framework=Terminus-22026.0256.260.7
GLM-5Agent Framework=Claude Code2026.0256.261.1
Gemini 3 ProAgent Framework=Terminus-22026.0254.2—
GPT-5.2 (xhigh)Agent Framework=Terminus-22026.0254—
Kimi K2.5Agent Framework=Terminus-22026.0250.8—
DeepSeek-V3.2Agent Framework=Claude Code2026.0246.4—
GLM-4.7Agent Framework=Terminus-22026.0241—
DeepSeek-V3.2Agent Framework=Terminus-22026.0239.3—
GLM-4.7Agent Framework=Claude Code2026.0232.8—