TopBench (Agentic) - Filtering (F1): leaderboard
Metric: F1 (0 to 1) on TopBench's filtering questions with classification targets, solved by the authors' ReAct agent that writes and runs Python in a Docker sandbox over the table: overlap between the set of records in the agent's CSV and the reference set of records whose predicted class meets the query; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash | 0.58 |
| 2 | Claude Sonnet 4.5 | 0.55 |
| 3 | DeepSeek V3.2 | 0.48 |
| 4 | Qwen 3 235B A22B 2507 (Thinking) | 0.42 |
| 5 | GPT-5.2 | 0.38 |
| 6 | DeepSeek V3.2 (Thinking) | 0.38 |
| 7 | Qwen 3 235B A22B 2507 Instruct | 0.28 |
Interactive version: theaggregate.ai/benchmark?slug=topbench-agentic-filtering-f1 · How It Works · Data refreshed daily, snapshot 2026-10-07.