XTC-Bench - Understanding: leaderboard
XTC-Bench (2026) visual understanding on 2,000 COCO and Visual Genome images: one question per scene-graph fact (object retrieval, attributes, relations), answers judged 0-5 against the reference. Understanding score (0-1, shown times 100); higher is better.
Metric: Overall understanding score (0-1, times 100) on visual questions derived one per fact from the reference scene graph of the real image, a Qwen3-235B judge scores each fact 0-5, normalized to 0-1 and shown times 100, over XTC-Bench's 2,000 scene-graph-annotated images from COCO 2017 val and Visual Genome (over 31,000 atomic facts about objects, attributes and relations); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 10 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Flash Image | 74 | |
| 2 | BAGEL-7B-MoT | 72.4 | #4619 |
| 3 | BLIP3-o8B | 69.2 | |
| 4 | GPT-5 | 68.4 | #178 |
| 5 | Janus-Pro-7B | 68.1 | #6857 |
| 6 | OmniGen2 | 65.9 | #4579 |
| 7 | Show-o2-7B | 60.5 | |
| 8 | Tar-7B | 58 | |
| 9 | Show-o | 50.3 | |
| 10 | MMaDA-8B | 38 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=xtc-bench-understanding · How It Works · Data refreshed daily, snapshot 2026-11-21.