ATRBench — leaderboard
Metric: TSAcc Default (%) (self-reported). Source: benchmarklist.com. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V4 Flash | 23.7 |
| 2 | Claude Opus 4.7 | 22.2 |
| 3 | MiniMax-M2.7 | 20.4 |
| 4 | Qwen 3.6 Plus | 20.3 |
| 5 | Gemini 3.1 Pro (Preview) | 20.1 |
| 6 | DeepSeek V4 Pro | 20 |
| 7 | Gemini 3 Flash (Preview) | 15.6 |
| 8 | GPT-5.4 | 15 |
Interactive version: theaggregate.ai/benchmark?slug=atrbench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.