Language Modeling
Benchmarks
DatasetSOTA methodMetricTrendResultsLast updated
2.73Perplexity
4,085
Jun 26, 2026
1.08Perplexity (PPL)
2,862
Jun 29, 2026
1.2PPL
2,416
Jul 9, 2026
1Perplexity
1,688
May 21, 2026
4.77Perplexity
1,565
May 15, 2026
4.345Perplexity
1,234
May 19, 2026
1.7PPL
908
Jul 3, 2026
2.22Perplexity
773
Jun 29, 2026
0.284PPL
740
May 13, 2026
8.159Perplexity
543
May 27, 2026
4.97Perplexity
488
Jun 11, 2026
5.52Perplexity
482
Jul 7, 2026
46.34Perplexity
450
Jul 7, 2026
2.58Perplexity
447
Jun 12, 2026
3.03Perplexity (PPL)
436
Jun 9, 2026
86.9Accuracy
412
May 29, 2026
1.7Perplexity
396
Jun 15, 2026
4.01PPL
382
Jun 23, 2026
2.9Word Perplexity
331
Jun 8, 2026
2Perplexity (PPL)
298
Apr 23, 2026
1.01PPL
290
Jun 23, 2026
1.906Validation Loss
259
Jun 30, 2026
3.1Perplexity
254
Jun 23, 2026
5Perplexity
244
Jun 15, 2026
4.59PPL
216
May 7, 2026