Atlas
Benchmarks
326 sources · 1,256 models × 2,956 benchmarks · 154,484 scoresScience's Last Exam: Scientific Computing
Whether an agent can fix scientific code to get scientifically valid results, scored as the share of 23 tasks passing every scientific check.
Models
6
Top score
57.0
Leader
Claude Opus 5
Updated
Nov 4, 2026
Resolution
—
Saturation
—
No signal or difficulty estimate. Too few models have been measured on this benchmark to estimate its difficulty.
| # | Model | Setting | Score | Source |
|---|---|---|---|---|
| 1 | Claude Opus 5 | high · Science's Last Exam; Claude Code with web search and fetch off; DiscoveryRSI for Hypothesis Discovery | 57.0 | |
| 2 | GPT-6.1 Sol | medium · Science's Last Exam; Codex with web search and fetch off; DiscoveryRSI for Hypothesis Discovery | 57.0 | |
| 3 | MiMo-V2.6-Pro | not reported · Science's Last Exam; Claude Code with web search and fetch off; DiscoveryRSI for Hypothesis Discovery | 30.0 | |
| 4 | Kimi K3.1 Pro | not reported · Science's Last Exam; Claude Code with web search and fetch off; DiscoveryRSI for Hypothesis Discovery | 26.0 | |
| 5 | MiniMax M3 | high · Science's Last Exam; Claude Code with web search and fetch off; DiscoveryRSI for Hypothesis Discovery | 22.0 | |
| 6 | GLM-5.3 Flash | high · Science's Last Exam; Claude Code with web search and fetch off; DiscoveryRSI for Hypothesis Discovery | 8.7 |
