Overreach Evaluation on OVEREAGER
4.8Composite-oracle Overreach RateGemini CLI
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini CLIModel=GPT-5.3-Codex2026.05 | 4.8 | |
| Claude CodeModel=GPT-5.3-Codex2026.05 | 5.4 | |
| Codex CLIModel=GPT-5.3-Codex2026.05 | 6.4 | |
| Gemini CLIModel=MiniMax-M22026.05 | 6.8 | |
| Claude CodeModel=GLM-52026.05 | 8 | |
| Gemini CLIModel=Sonnet-4.62026.05 | 11 | |
| Codex CLIModel=Sonnet-4.62026.05 | 11.2 | |
| Gemini CLIModel=Mean2026.05 | 11.2 | |
| Codex CLIModel=MiniMax-M22026.05 | 14 | |
| Claude CodeModel=Mean2026.05 | 14.6 | |
| Gemini CLIModel=GLM-52026.05 | 15 | |
| Codex CLIModel=Mean2026.05 | 16.08 | |
| Claude CodeModel=Gemini-2.5-Pro2026.05 | 16.8 | |
| Gemini CLIModel=Gemini-2.5-Pro2026.05 | 18.4 | |
| Claude CodeModel=Sonnet-4.62026.05 | 19.6 | |
| OpenHandsModel=GPT-5.3-Codex2026.05 | 22.6 | |
| Codex CLIModel=GLM-52026.05 | 23 | |
| Claude CodeModel=MiniMax-M22026.05 | 23.2 | |
| OpenHandsModel=Gemini-2.5-Pro2026.05 | 25.2 | |
| Codex CLIModel=Gemini-2.5-Pro2026.05 | 25.8 | |
| OpenHandsModel=MiniMax-M22026.05 | 33 | |
| OpenHandsModel=Mean2026.05 | 36.16 | |
| OpenHandsModel=Sonnet-4.62026.05 | 42.8 | |
| OpenHandsModel=GLM-52026.05 | 57.2 |