InfiniteScienceGym: leaderboard
Metric: Accuracy (%) on 500 templated questions (answerable and unanswerable) sampled from the first 500 simulated repositories, graded against the simulator's ground truth; models read files and run Python through MCP tools (proprietary models natively, open-weight models in a smolagents ReAct loop); higher is better. Source: arxiv.org. Saturation forecast: Around 2028. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 (Medium) | 49.2 |
| 2 | Claude Opus 4.6 (Medium) | 43.7 |
| 3 | GPT-OSS-20B | 29.7 |
| 4 | Qwen 3 4B 2507 Instruct | 26.5 |
| 5 | Gemma 3 27B (IT) | 24.2 |
Interactive version: theaggregate.ai/benchmark?slug=infinitesciencegym · How It Works · Data refreshed daily, snapshot 2026-10-07.