Agents' Last Exam v1.1
Created by UC Berkeley RDI, 14 Oct 2026 · Higher is better · Website
- Reported in a later post or card
Results
| Model | Score | Source | Source check |
|---|---|---|---|
| Gemini 4 Carbon28 Oct 2026 · Google | 77.4% | Gemini 4 Carbon: our next frontier model122 tasks; ALE-Claw harness; 5-hour window; October 17, 2026 task files; safety filters onSource image | |
| Gemini 4 Argon30 Sep 2026 · Google | 69.0% | Gemini 4 Carbon: our next frontier model122 tasks; ALE-Claw harness; 5-hour window; October 17, 2026 task files; safety filters on; reported laterSource image |