Benchmark
LABBench2
Science benchmarks measure how well models answer science questions. Scores are percentages. Higher is better.
- Models measured
- 3
- Scores recorded
- 5
- Sources
- 2
- IQ weight
- No weight published
Score distribution
- 85.1%
- Mean
- 86.2%
- Median
- 2.6%
- Std dev
- 82.1%
- 10th percentile
- 88.8%
- 90th percentile
- 82.1%
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Gemini 4 Argon· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 88.8% |
| 2 | Gemini 3.8 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86.2% |
| 3 | Gemini 3.8 Flashmacro-average across 11 subtasks· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 86.2% |
| 4 | Gemini 3.7 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 82.1% |
| 5 | Gemini 3.7 Flashoverall· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 82.1% |
Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats