Leaderboards
Expert-Driven Private Evaluations
Private Datasets
Scale’s proprietary, private evaluation datasets can’t be gamed, ensuring unbiased and uncontaminated results.
Evolving Competition
We periodically update leaderboards with new datasets and models, fostering a dynamic, contest-like environment.
Expert Evaluations
Our evaluations are performed by thoroughly vetted experts using domain specific methodologies, ensuring the highest quality and credibility.
Private Datasets
Scale’s proprietary, private evaluation datasets can’t be gamed, ensuring unbiased and uncontaminated results.
Evolving Competition
We periodically update leaderboards with new datasets and models, fostering a dynamic, contest-like environment.
Expert Evaluations
Our evaluations are performed by thoroughly vetted experts using domain specific methodologies, ensuring the highest quality and credibility.
Learn more about our evaluation methodology here →
Coding→
Learn More
Model | Score | 95% Confidence |
|---|---|---|
1176 | +34/-36 | |
1146 | +25/-27 | |
3rd | 1138 | +31/-29 |
1095 | +27/-30 | |
1060 | +26/-26 | |
1026 | +28/-26 | |
996 | +26/-27 | |
981 | +27/-27 | |
980 | +24/-26 | |
915 | +26/-25 | |
790 | +30/-31 | |
699 | +36/-38 |
Instruction Following→
Learn More
Model | Score | 95% Confidence |
|---|---|---|
90.80 | +1.50/-1.60 | |
2nd | 88.60 | +1.50/-1.40 |
87.60 | +1.40/-1.30 | |
85.50 | +1.80/-1.70 | |
85.40 | +1.50/-1.60 | |
85.20 | +1.70/-1.80 | |
84.70 | +1.60/-1.60 | |
83.30 | +1.70/-1.80 | |
82.90 | +2.00/-2.00 | |
82.60 | +1.90/-1.90 | |
73.60 | +2.40/-2.30 | |
66.80 | +2.10/-2.20 |
Math→
Learn More
If you’d like to add your model to this leaderboard or a future version, please contact seal@scale.com. To ensure leaderboard integrity, we require that models can only be featured the FIRST TIME when an organization encounters the prompts.