LatentMD - Content Correct: leaderboard
Metric: Content-correct responses (%). Source: arxiv.org. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4 | 93 |
| 2 | GPT-4o | 80 |
| 3 | Gemini 2.5 Flash | 76.1 |
| 4 | Gemma 3 27B | 76 |
| 5 | Qwen 3 32B | 71.6 |
| 6 | Llama 3.1 70B | 69.1 |
| 7 | Llama 3.1 8B | 48.2 |
| 8 | Gemma 2 9B | 15.5 |
Interactive version: theaggregate.ai/benchmark?slug=latentmd-content-correct · How It Works · Data refreshed daily, snapshot 2026-09-20.