LiveBench Connections — leaderboard
Metric: Score. Source: livebench.ai. 73 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o (2024-08-06) | 58 |
| 2 | GPT-4o (2024-05-13) | 53 |
| 3 | GPT-4o ChatGPT | 51.33 |
| 4 | Claude 3.5 Sonnet (20240620) | 49.67 |
| 5 | GPT-4 Turbo | 44.17 |
| 6 | Llama 3.1 405B Instruct | 44.17 |
| 7 | GPT-4 Preview (0125) | 43.17 |
| 8 | Gemini 1.5 Pro (Preview 0827) | 42.5 |
| 9 | Llama 3.1 70B Instruct | 41.67 |
| 10 | GPT-4 (0613) | 39.17 |
| 11 | Claude 3 Opus (20240229) | 37 |
| 12 | Gemini 1.5 Pro (Preview 0801) | 37 |
| 13 | GPT-4o Mini (2024-07-18) | 36.5 |
| 14 | Qwen 2 72B Instruct | 34 |
| 15 | Hermes 3 - Llama-3.1 70B | 32 |
Interactive version: theaggregate.ai/benchmark?slug=livebench-connections · How the rankings work · Data refreshed daily, snapshot 2026-07-22.