CodeMixQA: leaderboard
Metric: SimpleQA F1 (%), mean of 4 code-switching perturbations. Source: arxiv.org. Saturation forecast: Around December 2026. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4.1 | 36.83 |
| 2 | GPT-OSS-120B | 22.23 |
| 3 | Qwen 3 30B A3B (Thinking) | 20.84 |
| 4 | Llama 3.3 70B Instruct | 17.21 |
| 5 | GPT-OSS-20B | 16.48 |
| 6 | Gemma 3 27B | 11.38 |
| 7 | Qwen 3 4B | 8.38 |
| 8 | Gemma 3 12B | 8.1 |
Interactive version: theaggregate.ai/benchmark?slug=codemixqa · How It Works · Data refreshed daily, snapshot 2026-09-25.