SciUniverse - Optimize reactions with limited material: leaderboard

Task family in C5R's SciUniverse Level 1 (2026-09-24): with scarce target material, spend 24 simulated experiments over two plates on sixteen related amide couplings (recorded AIChemEco outcomes), then nominate a condition for the target. Passes when its confirmation yield closes at least 90% of the gap from a random-search baseline to the task-constrained oracle. 12 tasks, 4 attempts each. Pass@1 (%), higher is better.

Metric: Pass@1 (%). Source: c5r.net. Saturation forecast: Not forecast. 6 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 3.8 Flash (High)18.8#36 (Gemini 3.8 Flash)
2GPT-5.6 Sol (xHigh)16.7#29 (GPT-5.6 Sol)
3Claude Opus 5 (xHigh)14.6#20 (Claude Opus 5)
4Claude Fable 5.1 (xHigh)14.6#14 (Claude Fable 5.1)
5GPT-6 Astra (xHigh)8.3#6 (GPT-6 Astra)
6Grok 4.6 (xHigh)8.3#44 (Grok 4.6)

No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.

Interactive version: theaggregate.ai/benchmark?slug=sciuniverse-optimize-reactions-with-limited-material · How It Works · Data refreshed daily, snapshot 2026-11-12.