SCICONVBENCH: leaderboard
Large Language Models (LLMs) are increasingly deployed as scientific AI as- sistants, and a growing body of benchmarks evaluates their capabilities across knowledge retrieval, reasoning, code generation, and tool use. These evaluations, however, typically assume the scientific problem is already well-posed, whereas practical scientific assistance often begins with an ill-posed user request that mu
Metric: FRR (self-reported). Source: benchmarklist.com. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 94.7 | #152 |
| 2 | Gemini 2.5 Flash | 89 | #245 |
| 3 | GPT-5.2 | 84.1 | #110 |
| 4 | Claude Sonnet 4.6 | 75.9 | #90 |
| 5 | GPT-OSS-120B | 72.6 | #341 |
Interactive version: theaggregate.ai/benchmark?slug=sciconvbench · How It Works · Data refreshed daily, snapshot 2026-10-20.