MARCA (English, Basic): leaderboard
Metric: Checklist accuracy (0-1, times 100): share of a question's checklist items (expected entities and attributes) that a GPT-4.1 judge marks satisfied, averaged over MARCA's 52 English questions and three runs, Basic framework (the model itself calls web_search via the Serper API and web_scrape); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 88.5 |
| 2 | GPT-5.2 | 83.7 |
| 3 | GPT-5 Mini | 83.7 |
| 4 | Gemini 3 Flash | 82.8 |
| 5 | GPT-4.1 | 81.5 |
| 6 | Kimi K2 | 76.9 |
| 7 | Qwen 3 235B A22B | 66.8 |
| 8 | Gemini 2.5 Pro | 65.8 |
| 9 | GPT-4.1 Mini | 64 |
| 10 | Gemini 2.5 Flash | 52.8 |
| 11 | Qwen 3 30B A3B | 46.9 |
| 12 | Gemini 2.5 Flash Lite | 26 |
Interactive version: theaggregate.ai/benchmark?slug=marca-english-basic · How It Works · Data refreshed daily, snapshot 2026-10-07.