BBH
BIG-Bench Hard · a curated subset of 23 challenging tasks from BIG-Bench where language models previously failed to outperform average humans.
The Frontier
Best score over time · one chart, every benchmark
Full rankings
24 models tested · sorted by score
| # | Model | Score |
|---|---|---|
| 1 | 83.3 | |
| 2 | 78.7 | |
| 3 | 77.2 | |
| 4 | 75.2 | |
| 5 | 73.1 | |
| 6 | 72.1 | |
| 7 | 71.7 | |
| 8 | 66.8 | |
| 9 | 62.3 | |
| 10 | U Stable Beluga 2 | 59.1 |
| 11 | 48.8 | |
| 12 | 45.9 | |
| 13 | U Nemotron-4 15B | 44.9 |
| 14 | 44.3 | |
| 15 | 41.5 | |
| 16 | 40.0 | |
| 17 | U Yi 6B | 29.6 |
| 18 | U Baichuan 2-7B | 22.1 |
| 19 | U MPT-30B | 17.3 |
| 20 | 17.2 | |
| 21 | 16.1 | |
| 22 | 13.6 | |
| 23 | U INTELLECT-1 | 13.1 |
| 24 | U Baichuan1-7B | 10.0 |
Score distribution
Where models cluster
Correlated benchmarks
Pearson r · original research
Benchmarks that track with BBH
Pearson correlation across models scored on both benchmarks. Closer to 1 = strongly predictive.
Frequently asked
About BBH
What does BBH measure?
BIG-Bench Hard · a curated subset of 23 challenging tasks from BIG-Bench where language models previously failed to outperform average humans. 24 AI models have been tested on it. Scores range from 10.0 to 83.3 out of 100.
Which model leads on BBH?
DeepSeek V3 from DeepSeek leads BBH with a score of 83.3. The median score across 24 tested models is 45.4.
Is BBH saturated?
No · the top score is 83.3 out of 100 (83%). There is still meaningful room for improvement on BBH.
Does BBH predict performance on other benchmarks?
Yes · BBH scores correlate 0.98 with CMMLU across 5 shared models. Models that do well on BBH tend to do well on CMMLU.
How often is BBH data refreshed?
BenchGecko pulls updates daily. New model scores on BBH appear as soon as they are published by Epoch AI or the model provider.
- Category
- Reasoning
- Max score
- 100
- Models
- 24
- Updated
- 2024-12-26
Top on BBH
DeepSeek V3 · 83.3Gemini 1.5 Pro (Feb 2024) · 78.7Llama 3.1 405B · 77.2phi-3-medium 14B · 75.2Qwen2.5 72B Instruct · 73.1More reasoning benchmarks
Same category · related evaluations