FlatSeek Study - Literature: leaderboard
The FlatSeek study tests untouched symbolic-regression baselines on 83 literature equations and 200 synthetic equations, allowing up to 64 candidate expressions. Scores are counts of oracle-accepted expressions, not percentages. The paper’s trained FlatSeek adapter is a distinct artifact. This board reports systems with an accepted candidate (count out of 83). Higher is better.
Metric: Systems with an accepted candidate (count out of 83). Source: arxiv.org. Saturation forecast: Around February 2027. 5 models tracked. Updated 14 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V4 Flash | 67 | #259 |
| 2 | Qwen 3.5 27B | 61 | #365 |
| 3 | Qwen 3.5 9B | 59 | #814 |
| 4 | DeepSeek R1 Distill Qwen 32B | 54 | #1822 |
| 5 | Qwen 3.5 2B | 50 | #5184 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=flatseek-study-literature · How It Works · Data refreshed daily, snapshot 2026-11-16.