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

Text-to-SQL research benchmarks – Page 4 · SOTA2 Research

Text-to-SQL

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Synthesized Business Logic-Driven Dataset

90.24Single Execution Accuracy

9

Feb 26, 2026

Spider SmBoP parser outputs (dev)

78EM

9

Feb 26, 2026

Spider BRIDGEv2 parser outputs (dev)

72.5EM

9

Feb 26, 2026

Spider CodeT5 parser outputs (dev)

69.2EM

9

Feb 26, 2026

Spider-Realistic 1.0 (test)

RESDSQL-3B + NatSQL

77.4Exact Match (EM)

9

Feb 26, 2026

WikiSQL ADVETA-ADD

79.9Exact Match (EM)

9

Feb 26, 2026

WikiSQL ADVETA-RPL

69.2Exact Match (EM)

9

Feb 26, 2026

WikiSQL original (dev)

84.3Exact Match (EM)

9

Feb 26, 2026

WikiSQL Weakly-supervised (test)

GRAPPA (MLM+SSP)

84.7Accuracy

9

Feb 26, 2026

WikiSQL Weakly-supervised (dev)

GRAPPA (MLM+SSP)

85.9Accuracy

9

Feb 26, 2026

Spider2-Lite Snowflake

60.08Pass@1

8

May 22, 2026

Spider2-Lite SQLite

62.59Pass@1 Accuracy

8

May 22, 2026

0.65Spearman Correlation

8

Apr 7, 2026

0.74Spearman Correlation

8

Apr 7, 2026

0.46Spearman Correlation

8

Apr 7, 2026

MIMICSQL CDB (test)

94.2Execution Accuracy

8

Mar 9, 2026

82.06Accuracy (Easy)

8

Feb 26, 2026

25.1EM Accuracy

8

Feb 26, 2026

WTQ (ADVETA-ADD)

44.6Exact Match (EM)

8

Feb 26, 2026

WTQ (ADVETA-RPL)

41.8Exact Match (EM)

8

Feb 26, 2026

WTQ original (dev)

44.1Exact Match (EM)

8

Feb 26, 2026

74Accuracy

8

Feb 26, 2026

Content Enhanced BERT-based Text-to-SQL Generation

84.5Logic Form Accuracy

8

Feb 26, 2026

Content Enhanced BERT-based Text-to-SQL Generation

85.4Logic Form Acc

8

Feb 26, 2026

WikiSQL 1.0 (test)

82.6Execution Accuracy

8

Feb 26, 2026

Page 4 of 6

Popular datasets

Spider BIRD WikiSQL WTQ ATIS

Follow for updates

Join SOTA2 Contact