multimodal benchmark
BLINK
BLINK: Multimodal Large Language Models Can See but Not Perceive. A benchmark for multimodal language models focusing on core visual perception abilities. Reformats 14 classic computer vision tasks into 3,807 multiple-choice questions paired with single or multiple images and visual prompting. Tasks include relative depth estimation, visual correspondence, forensics detection, multi-view reasoning, counting, object localization, and spatial reasoning that humans can solve 'within a blink'.
Updated Aug 11, 2026
Models13
Model coverage13
MetricScore
EvidenceB
BLINK Ranking
Higher score ranks better on this benchmark.
| 01 | BY | 81.4% | 100.0% | 13 | C | |
| 02 | BY | 79.4% | 91.7% | 13 | C | |
| 03 | AC | 70.7% | 83.3% | 13 | C | |
| 04 | AC | 69.1% | 75.0% | 13 | C | |
| 05 | AC | 68.7% | 66.7% | 13 | C | |
| 06 | AC | 68.5% | 58.3% | 13 | C | |
| 07 | AC | 67.7% | 50.0% | 13 | C | |
| 08 | AC | 67.3% | 41.7% | 13 | C | |
| 09 | AC | 67.1% | 33.3% | 13 | C | |
| 10 | AC | 65.8% | 25.0% | 13 | C | |
| 11 | AC | 65.4% | 16.7% | 13 | C | |
| 12 | AC | 63.4% | 8.3% | 13 | C | |
| 13 | MI | 61.3% | 0.0% | 13 | C |
BLINK Score Distribution
A closer view of the leading scores on this benchmark.
BLINK Highlights
The leading models and scores on this benchmark.
What is BLINK?
What BLINK measures and how its scores work.
BLINK: Multimodal Large Language Models Can See but Not Perceive. A benchmark for multimodal language models focusing on core visual perception abilities. Reformats 14 classic computer vision tasks into 3,807 multiple-choice questions paired with single or multiple images and visual prompting. Tasks include relative depth estimation, visual correspondence, forensics detection, multi-view reasoning, counting, object localization, and spatial reasoning that humans can solve 'within a blink'.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- BLINK
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- blink|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about BLINK.
Which model scores highest on BLINK?
Seed 2.1 Pro is currently ranked first with 81.4%.
What does BLINK measure?
BLINK: Multimodal Large Language Models Can See but Not Perceive. A benchmark for multimodal language models focusing on core visual perception abilities. Reformats 14 classic computer vision tasks into 3,807 multiple-choice questions paired with single or multiple images and visual prompting. Tasks include relative depth estimation, visual correspondence, forensics detection, multi-view reasoning, counting, object localization, and spatial reasoning that humans can solve 'within a blink'.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
13 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.