Text-to-SQL
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
64.7Non-Seed Domain Score
30
Jun 11, 2026
58.3SemanticScore
30
Jun 11, 2026
82.5Accuracy
28
Apr 23, 2026
26.7Execution Accuracy (Spider 2.0 Test)
27
May 20, 2026
69.1Accuracy
27
Mar 25, 2026
90Execution Accuracy (EX)
24
Mar 18, 2026
1.07AVGLEN
22
Mar 16, 2026
56.1Gre Score
22
Feb 26, 2026
4.4Llama-8B Execution Accuracy
21
May 20, 2026
31.8Execution Accuracy (Llama-8B)
21
May 20, 2026
60.51Execution Accuracy (EX)
21
Jun 5, 2026
63.8Execution Accuracy (EX)
21
Apr 21, 2026
45.5Execution Accuracy
20
Jun 15, 2026
0.841Execution Accuracy
20
Jun 15, 2026
61.4Accuracy
20
May 27, 2026
83Exact Match Accuracy
17
Feb 26, 2026
63.1Execution Accuracy (SQLite)
16
May 11, 2026
88.4Accuracy (Total)
15
Jul 3, 2026
81.4Average Accuracy
15
May 7, 2026
6.67Match Accuracy (MAT)
15
Apr 24, 2026
64EA (%)
14
Feb 26, 2026
53.2Execution Accuracy
14
Feb 26, 2026
48.92Execution Accuracy (EX)
14
Feb 26, 2026
74.4Execution Accuracy (EX)
13
Jul 3, 2026
47.8Exact Match Accuracy
13
Feb 26, 2026