LVBench — leaderboard

LVBench evaluates model capability on multimodal tasks from the linked upstream source with Score as the primary reported metric.

Metric: Score (self-reported). Source: benchmarklist.com. Status: saturation imminent. 27 models tracked.

Top models

#ModelScore
1GPT-5.477.4
2Qwen 3.7 Plus76.2
3Gemini 3.1 Pro (Preview)75.1
4Qwen 3.6 Plus74.8
5Claude Opus 4.6 (Max)63
6GPT-4o (2024-11-20)48.9
7InternVL2.5-78B43.6
8Qwen 2 VL 72B41.3
9Gemini 1.5 Pro33.1
10GPT-4o (2024-05-13)30.8
11GPT-4o27

Interactive version: theaggregate.ai/benchmark?slug=lvbench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.