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

Instruction Following research benchmarks – Page 4 · SOTA2 Research

Instruction Following

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

ALFRED (test-unseen)

94.5GC

31

May 26, 2026

Staged Decoding

61.97IF-RT Score

30

Jun 1, 2026

76.26IF-RT Score

30

Jun 1, 2026

FollowRAG Instruction v1 (test)

41.1FollowRAG Instruction Score

30

Feb 26, 2026

FollowRAG Instruction

42.3FollowRAG Instruction Score

30

Feb 26, 2026

Claude-Opus-4.7

66.1CSR

29

May 11, 2026

Claude-Opus-4.7

81Average Score

29

May 11, 2026

Alpaca poisoned (test)

99.93F1 Score

28

May 13, 2026

AlpacaFarm Eval (test)

76.13Win Rate

28

Feb 26, 2026

IFEval v1 (test)

79.85Accuracy

28

May 27, 2026

64.23Avg@k

27

May 13, 2026

59.2Win Rate

27

Apr 6, 2026

WildBench (test)

58.6Info Seek

27

Feb 26, 2026

Super-Natural Instructions (test)

68.5ROUGE-L

27

Mar 5, 2026

AlpacaEval High-Variance (Top 20%) 2.0

rDPO (Best-of-K)

11.6Reward Score

26

Mar 10, 2026

AlpacaEval 2.0 (Overall)

rDPO (Best-of-K)

11.62Reward

26

Mar 10, 2026

69.3Score

26

Jul 2, 2026

86.5IFEval Score

25

May 25, 2026

7-benchmark suite SFT on Alpaca

71.5Accuracy

25

May 13, 2026

Inverse IFEval, MARS-Bench, MultiChallenge, Collie-Hard, EIFBench

99Pass@1

25

Mar 24, 2026

Tulu3 Evaluation Suite pool (test)

92.54ARC

25

Feb 26, 2026

WizardLM (test)

Llama-2-7B-star_instruct

6.87Score

25

Apr 29, 2026

DollyEval 500 (test)

31.3Rouge-L

24

Jun 30, 2026

FollowBench (test)

85.1CSR

24

Jun 4, 2026

ComplexBench (test)

79.8CSR

24

Jun 4, 2026

Page 4 of 19

Popular datasets

IFEval AlpacaEval IFBench MT-Bench Alpaca

Follow for updates

Join SOTA2 Contact