TableGPT2-7B [Su et al. '24]: benchmark results
Provider: Other.
Aggregate Rating 1529 ± 40, rank #1956 of 10916 rated models, from 4 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| TableBench | 27.95 | DP Score (%) | 45.2 |
| ReasonTabQA - English | 45.16 | Answer Accuracy (%; LLM judge against gold answers, program- | 34.4 |
| ReasonTabQA | 42.05 | Answer Accuracy (%; LLM judge against gold answers, program- | 31.2 |
| ReasonTabQA - Chinese | 39.81 | Answer Accuracy (%; LLM judge against gold answers, program- | 28.1 |
Interactive version: theaggregate.ai/model?slug=tablegpt2-7b-su-et-al-24 · How It Works · Data refreshed daily, snapshot 2026-11-08.