Journalistic Bias Accuracy: leaderboard
Journalistic Bias accuracy evaluates classification of media bias labels in news-style examples.
Metric: Accuracy (%). Source: github.com. Saturation forecast: Around 2034. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 44.44 | #343 |
| 2 | Claude Opus 4.7 | 38.78 | #49 |
| 3 | Gemma 4 31B (IT) | 35.88 | #176 |
| 4 | Llama 3.3 70B Instruct | 31.82 | #534 |
| 5 | GPT-4o Mini | 30.3 | #599 |
| 6 | GPT-5 (Medium) | 28.79 | #98 (GPT-5) |
| 7 | DeepSeek V3.1 | 27.27 | #264 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=journalistic-bias-accuracy · How It Works · Data refreshed daily, snapshot 2026-10-20.