ResearchGym: leaderboard
Autonomous AI-research-agent benchmark spanning the full ML research cycle across five realistic research tasks.
Metric: Best-of-3 score normalised to the task's reference (SOTA) result, SOTA = 100. Source: anikethh.github.io. Saturation forecast: Around March 2028. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 Codex (xHigh) | 62.1 | #95 (GPT-5.2 Codex) |
| 2 | Claude Opus 4.5 | 24 | #83 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=researchgym · How It Works · Data refreshed daily, snapshot 2026-10-20.