Aggregate
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Aggregate (IFEval, GPQA, LCB, Arena-Hard, CW, MT-Bench, WildBench)
63.69Average Score
14
Aggregate OBQA, CSQA, SIQA, ARC, MMLU, GSM8K-MC, AQUA
86.21Average Accuracy
14
Aggregate K&R, IFEval-PT, HumanEval
53.64Average Score
14
Aggregate (CIFAR-100, CIFAR-100C, SVHN, Places365, Attacks)
96Mean Score
14
Aggregate (multiple datasets and shifts) (test)
0.064MAE
14
Aggregate IOI, MCQA, Arithmetic, ARC
87CPR Faithfulness Score
13
Aggregate ID/OOD
37.3ID Score
13
Aggregate Image and Video Benchmarks
100RelAcc
13
Aggregate (Cola, GEmot, IMDB, News, SST5, Toxigen, YELP)
8.4ECE
13
Aggregate Math, Code, Science
1,267Response Length (Tokens)
12
Aggregate (FB15K, FB15K-237, WN18, WN18RR, NELL995, DDB14) (test)
1Entity Distance (ED)
12
Aggregate All evaluated datasets (test)
48.4Precision@1
12
Aggregate AIME, AMC, GSM8K, MATH, MNV, OLPD
173Avg Training Steps to Target Acc
12
Aggregate AIME-2024 AIME-2025 MATH500 OlympiadBench AMC23
52.57Accuracy
12
Aggregate Debt, Warfarin, TCGA, News, NewsHet
2.2Average Rank
12
Aggregate (Derm7pt, PAD, PH2) (out-of-distribution)
84.11Avg ROC-AUC
12
Aggregate MATH500, AMC, OlympiadBench, AIME24, AIME25
66.6Average Score
11
Aggregate VSI-Bench, VideoMMMU, MMVU, MVBench, TempCompass, VideoMME
54.1Average Score
11
Aggregate IFEval, IFBench, Arena-Hard-v2.0, Creative Writing v3, WritingBench
71.9Average Score
11
Aggregate Wino Hella ARC-E ARC-C MMLU
56.67Average Accuracy
11
Aggregate VOC, Context, ADE, Cityscapes, COCO Stuff
39Average mIoU
11
Aggregate (GSM8K, MATH500, Minerva Math, Olympiad Bench, AIME24, AIME25, GPQA)
79.3Accuracy
11
Aggregate 6 Benchmarks Average
62.04Average Accuracy
10
Aggregate HealthBench, LLMMed-Eval, WritingBench, Creative Writing, ResearchQA
70.55Macro-average Score
10
Aggregate MATH500, AMC23, AIME24, AIME25, BBEH
44.8Average Score
10