SPUR - Heterogeneous Integration: leaderboard
SPUR (2026) tests multimodal models with multiple-choice questions on multi-panel biomedical experimental figures from PubMed Central. Accuracy in percent on heterogeneous integration questions, which ask the model to combine evidence across panels of different types; higher is better.
Metric: Accuracy (%) on SPUR the 130 heterogeneous integration multiple-choice questions about multi-panel biomedical experimental figures from PubMed Central papers (questions drafted by GPT-4o, kept only when GPT-4o failed them in at least six of ten text-only attempts, then expert-reviewed): the questions ask the model to align and reason across panels of different types; higher is better. Source: arxiv.org. Saturation forecast: Not forecast. 19 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Ministral-3-14B-Instruct-2512 | 70 | #1639 | |
| 2 | GLM-4.5V | 68.46 | #754 | |
| 3 | Ministral-3-8B-Instruct-2512 | 66.15 | #1907 | |
| 4 | Qwen 3 VL 30B A3B Instruct | 63.08 | #895 | |
| 5 | Qwen 2.5 VL 72B Instruct | 61.9 | #841 | |
| 6 | Gemini 2.5 Pro (Preview 06-05) | 61.54 | #213 | |
| 7 | InternVL3-78B | 61.24 | #759 | |
| 8 | Qwen 3 VL 30B A3B (Thinking) | 61.24 | #719 | |
| 9 | Claude 3.7 Sonnet | Thinking | 60.8 | #470 |
| 10 | Gemini 3 Pro (Preview) | 59.23 | #119 | |
| 11 | O4 Mini | High | 59.23 | #368 |
| 12 | Llama 4 Maverick | 58.46 | #1110 | |
| 13 | Gemma 3 27B (IT) | 57.69 | #1324 | |
| 14 | Mistral Small 3.1 | 56.92 | #1706 | |
| 15 | Seed-1.6 | 56.92 | #469 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=spur-heterogeneous-integration · How It Works · Data refreshed daily, snapshot 2026-11-21.