Open LMM Reasoning - LogicVista - spatial — leaderboard

Metric: Accuracy (%). Source: huggingface.co. 83 models tracked.

Top models

#ModelScore
1GPT-4.1 Mini34.6
2InternVL3-38B34.6
3Doubao-1.5-Pro34.6
4Gemini 2.0 Flash33.3
5Qwen 2 VL 72B30.8
6Claude 3.7 Sonnet30.8
7GPT-4.1 (2025-04-14)30.8
8QVQ-72B-Preview30.8
9Grok 2 (1212)29.5
10GPT-4o ChatGPT29.5
11Gemini 1.5 Pro (002)28.2
12Llama 3.2 11B Instruct26.9
13Qwen 2 VL 2B25.6
14Claude 3.5 Sonnet (20241022)25.6
15InternVL3-78B24.4

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