TopBench (Agentic) - Single-Point Prediction: leaderboard
Metric: Prediction accuracy (0 to 1) on TopBench's 274 single-point questions, which ask for an unobserved outcome of a profile given a historical table, solved by the authors' ReAct agent that writes and runs Python in a Docker sandbox over the table: a regression answer earns 0.6 x (1 - range-normalized absolute error) plus 0.4 x the overlap of its predicted interval with the reference interval, penalized when the interval is too wide, and a classification answer earns credit for the right class; values and decisions extracted from the free-text answer by a DeepSeek-V3.2 judge and checked against the answer text; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash | 0.66 |
| 2 | Claude Sonnet 4.5 | 0.64 |
| 3 | DeepSeek V3.2 (Thinking) | 0.61 |
| 4 | GPT-5.2 | 0.6 |
| 5 | DeepSeek V3.2 | 0.58 |
| 6 | Qwen 3 235B A22B 2507 (Thinking) | 0.57 |
| 7 | Qwen 3 235B A22B 2507 Instruct | 0.43 |
Interactive version: theaggregate.ai/benchmark?slug=topbench-agentic-single-point-prediction · How It Works · Data refreshed daily, snapshot 2026-10-07.