INCLUDE: leaderboard
INCLUDE multilingual evaluation reported in Anthropic's Claude Opus 4.8 system card.
Metric: Score (self-reported). Source: benchmarklist.com. Status: saturation imminent. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 90.7 |
| 2 | Claude Mythos 5 | 90.5 |
| 3 | Claude Mythos Preview | 90.1 |
| 4 | Claude Opus 5 | 89.8 |
| 5 | Claude Opus 4.8 | 88.5 |
| 6 | GPT-5.4 | 88.3 |
| 7 | Claude Opus 4.6 (Max) | 87.4 |
| 8 | Claude Opus 4.7 | 87 |
| 9 | Claude Sonnet 5 | 86.5 |
| 10 | Qwen 3.7 Max | 86.2 |
| 11 | DeepSeek V4 Pro (Max) | 86.1 |
| 12 | Claude Sonnet 4.6 | 85.7 |
| 13 | Qwen 3.6 Plus | 85.1 |
| 14 | GLM-5.1 (Thinking) | 84.3 |
| 15 | Kimi K2.6 (Thinking) | 84.2 |
Interactive version: theaggregate.ai/benchmark?slug=include · How It Works · Data refreshed daily, snapshot 2026-09-05.