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

Long-context Understanding research benchmarks – Page 4 · SOTA2 Research

Long-context Understanding

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

41.35LongBench 32k Score

9

Jun 16, 2026

Average MuSiQue, 2WikiMQA, MFQA-En, NarrativeQA, En.QA

42.48SubEM Score

9

Apr 10, 2026

17.5SubEM

9

Apr 10, 2026

28.33SubEM

9

Apr 10, 2026

39.17SubEM

9

Apr 10, 2026

40SubEM

9

Apr 10, 2026

ZeroSCROLLS (val)

50.5Qasper F1

9

Jun 29, 2026

LongChat-7B-v1.5

27.7Qasper Score

8

Jun 9, 2026

70.6Coursera Accuracy

6

Jun 29, 2026

LongBench 16K context

19.02LongBench Score (16k)

6

Jun 16, 2026

45.77Qasper

6

Mar 13, 2026

InfiniteBench (test)

36.7En QA F1

6

Feb 26, 2026

LongBench-Zh filtered (test)

Sentinel (Qwen2.5-0.5B-Instruct)

64.8Single Document Score

5

Jun 15, 2026

LLaDA-8B-Instruct

100Subtask 1 Score (s1)

5

Jun 2, 2026

91.42RULER Score

5

May 21, 2026

RULER 32k tokens

100NIAH-S

5

May 11, 2026

RULER 128K context

48.38NIAH

5

Apr 15, 2026

68Accuracy

5

Apr 13, 2026

25GvR Score

5

Mar 30, 2026

RULER MRCR and NoLiMa

38.97Average Score

5

Feb 26, 2026

54.54Accuracy (32K Context)

5

Feb 26, 2026

92.65RULER Accuracy (64K Context)

5

Feb 26, 2026

NIH Multi-needle

100Accuracy

5

Feb 26, 2026

InfiniteBench En.MC

83.4Accuracy

5

Feb 26, 2026

89.6RULER 512K Score

4

Jun 9, 2026

Page 4 of 6

Popular datasets

LongBench RULER InfiniteBench LV-Eval HELMET

Follow for updates

Join SOTA2 Contact