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 evaluation research benchmarks – Page 2 · SOTA2 Research

Long-context evaluation

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

1.23Context Window Error (CWE)

6

Apr 20, 2026

Humanity's Last Exam AA-LCR

54.3Accuracy

6

Feb 26, 2026

Long Context Benchmarks

32.3MDQA-10 Score

5

Feb 26, 2026

RULER 16k context length

73.2Accuracy

4

Jul 3, 2026

RULER 8k context length

83.8Accuracy

4

Jul 3, 2026

RULER 128K sequences Llama3.1-70B-Instruct

65.03RULER Score

4

Apr 1, 2026

RULER ultra-long context official

Qwen3-Next-80B-A3B-Instruct

96Accuracy (128K)

4

Feb 26, 2026

100NS1 (Sequence Accuracy 1)

3

Apr 15, 2026

LongBench context-dependent transfer v2

36.4Score (32k Context)

2

Jun 9, 2026

RULER 32k context Average 13 tasks

0.635Score

2

May 15, 2026

RULER 16k context Average 13 tasks

75Score

2

May 15, 2026

RULER 8k context Average 13 tasks

79.3Score

2

May 15, 2026

RULER 4k context Average 13 tasks

0.842Score

2

May 15, 2026

Page 2 of 2

Popular datasets

RULER LongBench Humanity's Last Exam L-Eval Long Context Benchmarks

Follow for updates

Join SOTA2 Contact