VisuLogic
Seed2.1 Pro 54.3% leads since Jun 2026 Fading Models answer 1,000 visual reasoning puzzles in six categories that cannot be solved by describing the image in words. Scored by accuracy.
The paper reports human accuracy near 51 percent and most models near the 25 percent chance level.
In the authors’ words
Visual reasoning is a core component of human intelligence and a critical capability for advanced multimodal models. Yet current reasoning evaluations of multimodal large language models (MLLMs) often rely on text descriptions and allow language-based reasoning shortcuts, failing to measure genuine vision-centric reasoning. To address this, we introduce VisuLogic: a benchmark of 1,000 human-verified problems across six categories (e.g., quantitative shifts, spatial relations, attribute comparisons). These various types of questions can be evaluated to assess the visual reasoning capabilities of MLLMs from multiple perspectives. We evaluate leading MLLMs on this benchmark and analyze their results to identify common failure modes. Most models score below 30% accuracy-only slightly above the 25% random baseline and far below the 51.4% achieved by humans-revealing significant gaps in visual reasoning. Furthermore, we provide a supplementary training dataset and a reinforcement-learning baseline to support further progress.
- Level
- Everyday Most adults, without training
- Scored
- Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
- Answers
- Multiple choice
Seed2.1 Pro leads with 54.3% among models released by Jun 2026
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
The newest model with a score here came out in Jun 2026, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6 Astra
- Claude Opus 5.5
- GPT-6.1 Terra
- GPT-6.1 Sol
- and 16 more
- 1 Seed2.1 Pro 54.3%
- 2 Seed2.0 Pro 47.4%
- 3 Seed1.5-VL 35%
- 4
Qwen3-VL-235B-A22B 34.4% - 5
ERNIE 5.0 32% - 6 ERNIE-4.5-VL-28B-A3B-Thinking 31.1%
- 7
STEP3-VL-10B 29.7%
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Seed2.1 Pro | 54.3% | Jun 2026 | Model card | |
| 2 | Seed2.0 Pro | 47.4% | Feb 2026 | Model card | |
| 3 | Seed1.5-VL | 35% | May 2025 | Model card | |
| 4 | Qwen3-VL-235B-A22B | 34.4% | Sep 2025 | Model card | |
| 5 | ERNIE 5.0 | 32% | Feb 2026 | Model card | |
| 6 | ERNIE-4.5-VL-28B-A3B-Thinking | 31.1% | Nov 2025 | Model card | |
| 7 | STEP3-VL-10B | 29.7% | Jan 2026 | Model card |
The benchmark's own leaderboard has 25 entries, 17 of them models we track. We plot every source in the charts below.
3 models have held number one since its release in Apr 2025
Lab claims count from the day the card came out, independent runs from the model’s release. We draw one row each for the benchmark’s own leaderboard and what labs claim in their cards.
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Labs headline it less often now, last in Feb 2026
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| 14 Feb 2026 | Seed2.0 Pro | 47.4 | VisuLogic |
| 6 Feb 2026 | ERNIE 5.0 | 32 | VisuLogic |
| 11 Nov 2025 | ERNIE-4.5-VL-28B-A3B-Thinking | 31.1 | VisuLogic |
| 23 Sep 2025 | Qwen3-VL-235B-A22B | 34.4 | VisuLogic |
| 12 May 2025 | Seed1.5-VL | 35 | VisuLogic |
6 system cards report it, from 4 labs
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| 23 Jun 2026 | Seed2.1 Pro model card | 54.3 | VisuLogic (Table 9) | Capabilities |
| 14 Feb 2026 | Seed2.0 Pro model card | 47.4 | VisuLogic (Table 8) | Capabilities |
| 4 Feb 2026 | ERNIE 5.0 technical report | 32 | VisuLogic (Table 3) | Capabilities |
| 14 Jan 2026 | STEP3-VL-10B technical report | 29.7 | VisuLogic, SeRe (Tables 1 and 3). PaCoRe 32.70 | Capabilities |
| 26 Nov 2025 | Qwen3-VL-235B-A22B technical report | 34.4 | VisuLogic (Table 2) | Capabilities |
| 11 May 2025 | Seed1.5-VL technical report | 35 | VisuLogic, thinking (Table 6). Non-thinking 33.0 | Capabilities |
More in Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. GPT-5.1 85.4%
- MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. Kimi K3.1 Pro 84.1%
- CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
- MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
- ChartQA In use Models answer human-written and generated questions about chart images that need visual and arithmetic reasoning. Scored by relaxed accuracy with a 5 percent numeric tolerance.
Command A Vision 90.9% - DocVQA In use Models answer 50,000 questions about 12,000 scanned document images by reading their text and layout. Scoring is ANLS, a string similarity to reference answers. Seed1.5-VL 96.9%