LiveBench Table Join — leaderboard
Metric: Score. Source: livebench.ai. 73 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o ChatGPT | 45.3 |
| 2 | Claude 3 Opus (20240229) | 40.96 |
| 3 | Claude 3.5 Sonnet (20240620) | 40.22 |
| 4 | GPT-4o (2024-08-06) | 38.68 |
| 5 | Gemini 1.5 Pro (0514) | 38.42 |
| 6 | Llama 3.1 405B Instruct | 36.52 |
| 7 | GPT-4 Preview (0125) | 36.18 |
| 8 | GPT-4 Turbo | 35.96 |
| 9 | Llama 3.1 70B Instruct | 30.88 |
| 10 | DeepSeek Coder V2 | 30.76 |
| 11 | Gemini 1.5 Pro (Preview 0827) | 30.5 |
| 12 | GPT-4o (2024-05-13) | 29.22 |
| 13 | Gemini 1.5 Pro (Preview 0801) | 26.46 |
| 14 | Hermes 3 - Llama-3.1 70B | 26.34 |
| 15 | Llama 3 70B Instruct | 25.24 |
Interactive version: theaggregate.ai/benchmark?slug=livebench-table-join · How the rankings work · Data refreshed daily, snapshot 2026-07-22.