Evals for Every Language - Language lua: leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 | 45.14 |
| 2 | Gemini 3.1 Pro (Preview) | 44.03 |
| 3 | Gemini 3.1 Flash Lite | 43.37 |
| 4 | Gemini 2.5 Pro | 42.32 |
| 5 | Claude Sonnet 4.5 | 40.54 |
| 6 | GPT-5 | 40.42 |
| 7 | GPT-5.3 | 40.3 |
| 8 | GPT-5.5 | 40.24 |
| 9 | Gemini 2.5 Flash | 39.62 |
| 10 | Claude Sonnet 4.6 | 38.79 |
| 11 | GPT-5.1 | 38.1 |
| 12 | GPT-5.4 | 37.73 |
| 13 | Claude Opus 4.8 | 36.55 |
| 14 | Qwen 3.6 Plus | 34.78 |
| 15 | Grok 4.20 | 34.61 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-lua · How It Works · Data refreshed daily, snapshot 2026-09-05.