Text Generation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
5.29Latency (ms/tok)
8
Jun 2, 2026
5.64Latency (ms/tok)
8
Jun 2, 2026
5.44Latency (ms/tok)
8
Jun 2, 2026
6.34Latency (ms/tok)
8
Jun 2, 2026
5.17Latency (ms/tok)
8
Jun 2, 2026
0.014MAUVE Score
8
Jun 1, 2026
31.01Alast
8
Mar 10, 2026
30.58Alast
8
Mar 10, 2026
2.25Alast
8
Mar 10, 2026
2.51Alast
8
Mar 10, 2026
30.4Throughput (tokens/s)
8
Feb 26, 2026
8.41Perplexity (PPL)
7
May 13, 2026
43.34Perplexity
7
May 12, 2026
34.96Perplexity
7
May 12, 2026
33.22Perplexity
7
May 12, 2026
8MIA Score
7
May 12, 2026
50MIA
7
May 12, 2026
96.6Mauve
7
May 7, 2026
0.701BERTScore
7
Apr 16, 2026
0.677BERTScore
7
Apr 16, 2026
0.708BERTScore
7
Apr 16, 2026
0Disallowed Tokens Count
7
Mar 31, 2026
0.78Lexical Distance
7
Feb 26, 2026
15.5BLEU-4
7
Feb 26, 2026
2.09Avg. Tokens
7
Feb 26, 2026