language benchmark
Spider Leaderboard
Spider is a cross-domain semantic parsing and text-to-SQL dataset with 10,181 questions and 5,693 unique complex SQL queries across 200 databases and 138 domains.
Updated Sep 5, 2026
Models2
Model coverage2
MetricScore
EvidenceB
Spider Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelMA | Score63.50% | Percentile100.00% | Participants2 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score31.10% | Percentile0.00% | Participants2 | EvidenceC | Evaluated |
Spider Highlights
The leading models and scores on this benchmark.
Spider Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for Spider
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is Spider?
What Spider measures and how its scores work.
Spider is a language benchmark annotated by 11 college students that contains questions paired with complex SQL queries over databases with multiple tables.
It measures models' ability to generalize to new SQL queries and new database schemas.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- Spider
- Modality
- text
- Primary category
- language
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about Spider.
Which model scores highest on Spider?
Codestral-22B is currently ranked first with 63.50%.
What are the top three models on Spider?
The current leaders are Codestral-22B (63.50%) and Qwen3-Coder 480B A35B Instruct (31.10%).
Which Spider model has the lowest official input price?
Qwen3-Coder 480B A35B Instruct has the lowest matched official input price at $1.5 input / $7.5 output per 1M tokens.
Which models are fastest among Spider results?
No matched runtime record is currently available.
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.