Benchmark
Agents' Last Exam Benchmark
What is Agents' Last Exam?
Multi-step agentic benchmark using Claude Code harness. Tool Search disabled. Scores depend on harness, reasoning effort, context length, and timeout settings.
- Released
- 2025-01-23
- Evaluates
- coding agent
- Openness
- unknown
- Importer
- Model reports
- Reported scores
- 3
Source provenance
- Original evidence https://lastexam.ai/
Agents' Last Exam links
- Website https://lastexam.ai/
Agents' Last Exam results and reported scores
Model reports
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.
Which sources cite Agents' Last Exam?
- DeepSeek-V4.1-Flash DeepSeek, 2026-09-10
- Hy4 preview Tencent, 2026-08-28
- GLM-5.3-Flash Z.ai, 2026-08-27