LithoBench - Practical Application (Open-Ended): leaderboard
Metric: OPEN_J score (0-10) = 0.4 x ROUGE-L x 10 + 0.2 x BLEU-4 x 10 + 0.4 x the GPT-5.4 judge's overall score (0-10), comparing the answer with the reference answer, on the 1,200 open-ended practical application questions of LithoBench (practical assessments such as mapping, engineering and resource implications in GF-2 remote-sensing image patches); temperature 0, 1,024 output tokens; higher is better. Source: arxiv.org. Saturation forecast: Around 2030. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Kimi K2.5 | 3.95 |
| 2 | GLM-4.6V | 3.86 |
| 3 | Qwen 3.5 397B A17B | 3.86 |
| 4 | Grok 4 | 3.69 |
| 5 | Gemini 3 Pro | 3.49 |
| 6 | Qwen 3.5 9B | 2.88 |
| 7 | Claude Sonnet 4.6 | 2.82 |
| 8 | Qwen 3.5 27B | 1.84 |
Interactive version: theaggregate.ai/benchmark?slug=lithobench-practical-application-open-ended · How It Works · Data refreshed daily, snapshot 2026-10-07.