LiveBench Table Join: leaderboard
Each item shows two tables from a recent public dataset whose columns partly overlap, with the second table's names anonymized, and asks for a valid join mapping from the first to the second. Scoring is automatic: an F1 over the proposed column pairs, counting extra pairs as false positives and missing ones as false negatives. This board is that task's LiveBench score, higher is better.
Metric: Score. Source: livebench.ai. Saturation forecast: Around 2034. 71 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra (Max) | 58.37 | #6 (GPT-6 Astra) |
| 2 | GPT-6.1 Sol (Max) | 57.42 | #7 (GPT-6.1 Sol) |
| 3 | Claude Fable 5.1 (Max) | 56.15 | #9 (Claude Fable 5.1) |
| 4 | Claude Fable 5 (Max) | 56.13 | #16 (Claude Fable 5) |
| 5 | GPT-5.5 (xHigh) | 55.9 | #31 (GPT-5.5) |
| 6 | GPT-6.1 Sol (xHigh) | 55.25 | #7 (GPT-6.1 Sol) |
| 7 | GPT-6 Sol (Max) | 54.02 | #14 (GPT-6 Sol) |
| 8 | Gemini 3.1 Pro (Preview) (High) | 52.35 | #58 (Gemini 3.1 Pro (Preview)) |
| 9 | Claude Opus 5.5 (Max) | 52.33 | #3 (Claude Opus 5.5) |
| 10 | Claude Opus 5 (Max) | 51.96 | #12 (Claude Opus 5) |
| 11 | GPT-5.4 (xHigh) | 51.75 | #79 (GPT-5.4) |
| 12 | Claude Opus 4.8 (Max) | 51.08 | #37 (Claude Opus 4.8) |
| 13 | Qwen3.8-Flash-Next | 51.02 | #72 |
| 14 | Claude Opus 5.5 (xHigh) | 51 | #3 (Claude Opus 5.5) |
| 15 | GPT-5.4 Nano (xHigh) | 50.59 | #328 (GPT-5.4 Nano) |
No result here: #8 Gemini 4 Argon, #10 Muse Wonder Full.
Interactive version: theaggregate.ai/benchmark?slug=livebench-table-join · How It Works · Data refreshed daily, snapshot 2026-10-20.