CreditCardQA: leaderboard
Metric: Chain-of-Thought Accuracy (%). Source: arxiv.org. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 70.5 |
| 2 | GPT-5 | 70.4 |
| 3 | GPT-5 Mini | 70 |
| 4 | GLM-4.6 | 65 |
| 5 | Kimi K2 (Thinking) | 64.1 |
| 6 | DeepSeek R1 | 63.3 |
| 7 | Llama 3.3 70B | 62 |
| 8 | Mistral Small | 61.1 |
Interactive version: theaggregate.ai/benchmark?slug=creditcardqa · How It Works · Data refreshed daily, snapshot 2026-09-05.