VLRS-Bench - Causal Reasoning: leaderboard
Metric: Score (%) on the 125 Causal Reasoning (single image) questions of VLRS-Bench remote-sensing reasoning questions built from 11 public optical, multi-temporal and change-detection datasets with DSM, NIR, mask and metadata priors, drafted by GPT-5-chat and reviewed by models and nine RS experts; single-choice (5 options) and true/false items score 1 when correct, 8-option multiple-choice and fill-in-the-blank items score 1 when fully correct and 0.5 when partially correct (blanks matched by embedding similarity above 0.8); zero-shot with a standardized prompt; mean item score times 100; higher is better. Source: arxiv.org. Saturation forecast: Not forecast. 23 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o Mini | 42.8 | #588 |
| 2 | GPT-5 Chat | 42.4 | #233 |
| 3 | Gemini 3.1 Pro (Preview) | 42 | #54 |
| 4 | GPT-5.4 | 41.2 | #76 |
| 5 | GPT-4o (2024-11-20) | 37.6 | #369 |
| 6 | Llama 3.2 90B | 36.8 | #575 |
| 7 | Claude 3.5 Haiku | 30.8 | #553 |
| 8 | Qwen 2.5 VL 32B Instruct | 29.2 | #443 |
| 9 | Claude Opus 4.6 | 27.2 | #60 |
| 10 | GLM-4.5V | 26.8 | #339 |
| 11 | Qwen 2.5 VL 7B Instruct | 25.6 | #643 |
| 12 | Llama 3.2 11B | 23.2 | #1183 |
| 13 | Qwen 2.5 VL 72B Instruct | 21.6 | #364 |
| 14 | Gemini 2.5 Flash | 20 | #237 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=vlrs-bench-causal-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-11.