Text-to-SQL on Spider (Kendall's τ)
-0.78Kendall's τGPT-OSS-120
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-OSS-120Rank subset=All ranks2026.04 | -0.78 | -1 | |
| GPT-OSS-120Rank subset=Ranks ≥ 32026.04 | -0.79 | -1 | |
| GPT-OSS-20Rank subset=Ranks ≥ 32026.04 | -0.86 | -1 | |
| GPT-4o-miniRank subset=Ranks ≥ 32026.04 | -0.86 | -1 | |
| LLaMA3.3-70BRank subset=Ranks ≥ 32026.04 | -0.86 | -1 | |
| LLaMA3.3-70BRank subset=All ranks2026.04 | -0.87 | -1 | |
| GPT-OSS-20Rank subset=All ranks2026.04 | -0.91 | -1 | |
| GPT-4o-miniRank subset=All ranks2026.04 | -0.91 | -1 | |
| O3-miniRank subset=All ranks2026.04 | -0.91 | -1 | |
| O3-miniRank subset=Ranks ≥ 32026.04 | -0.93 | -1 | |
| LLaMA3.1-405BRank subset=All ranks2026.04 | -0.96 | -1 | |
| LLaMA3.1-405BRank subset=Ranks ≥ 32026.04 | -1 | -1 |