Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Code dataset and benchmarks · SOTA2 Research

Research Datasets

Code

Benchmarks

Task NameDataset NameSOTA MethodSort

Task NameDataset NameSOTA ResultTrendResults

Code Generation

Code

98.7Accuracy

242

ECG Classification

CODE-15% (test)

97Macro AUROC

36

Continual Learning

Code Domain-level stream

62.09AA

34

AI-generated text detection

Code

0.979AUC

24

Speculative Decoding

Code

138.72Throughput (tokens/s)

22

Code Generation

Code

65.5Code Avg

18

Code HumanEval+ LiveCodeBench v5

79.88HEval+ (Pass@1)

18

Code Generation

Code Category Average (test)

77.43Accuracy

18

Incremental BPE Tokenization

Code

1.369End-to-end CPU Time (s)

15

ECG Interpretation

CODE 15%

86.4AUC

15

Code Generation

Code R1

19.96Pass@1

14

Code Generation

Code R1

9.55Score

14

Code Generation

Code

52.67Performance Score

12

BPE Tokenization

Code

2.88Speedup Factor

12

Machine-generated text detection

Code Llama-3-70B-Instruct (test)

0.951AUC

12

AI-generated text detection

Code GPT-3.5 Turbo

0.906AUC

12

Code Generation and Understanding

Code Crux, MultiPL_E, MBPP

60.12Crux Score

12

Code Generation

Code latest (test)

83.9HumanEval

12

Agentic Routing

Code MBPP HumanEval

76Accuracy

10

Answerability Prediction

CODE n=30 (matched pairs)

85AUC

9

ECG Interpretation

CODE (test)

96.79AUC

9

Multi-turn conversation performance

Code

98.3Avg Performance

9

Speculative Decoding

Code Stack-Edu

2.61Speedup

8

Code Generation

Code Out-of-Domain

63.99Accuracy

8

ECG Abnormality Detection

CODE 15%

91.5AUC

8

Page 1 of 2

PreviousNext

Popular tasks

Code Generation Speculative Decoding Language Modeling ECG Classification AI-generated text detection

Follow for updates

Join SOTA2 Contact