Text Generation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
37.5ROUGE-1
15
May 14, 2026
99.88TPR @ FPR=1%
15
Feb 26, 2026
1,508Elo Rating
14
Jul 7, 2026
4.76Speedup
14
Jun 2, 2026
41.97CIDEr
14
Mar 12, 2026
24Quality Better Count
14
Feb 26, 2026
32.15ROUGE-1
13
Mar 9, 2026
3.77Average Perplexity
13
May 25, 2026
34.2TemporalBench Score
12
Jun 30, 2026
75.4MVBench Score
12
Jun 30, 2026
58.083BLEU-4
12
Jun 30, 2026
34.6ROUGE-1
12
May 18, 2026
85.7BS
12
May 18, 2026
15.36PPL
12
Feb 26, 2026
55.6SWITCH Score
11
Jun 30, 2026
52.13DSRBench Score
11
Jun 30, 2026
0.94MAUVE
11
May 19, 2026
66.2BS
11
May 18, 2026
49.3RougeL
11
Apr 21, 2026
0.944Diversity
11
Mar 19, 2026
9.68BLEU
11
Feb 26, 2026
10.46BLEU
11
Feb 26, 2026
16.96BLEU
11
Feb 26, 2026
111.89Throughput (tokens/s)
10
May 4, 2026
117.17Throughput (tokens/s)
10
May 4, 2026