multimodal benchmark
BLINK Leaderboard
BLINK is a multimodal benchmark that evaluates core visual perception abilities using 3,807 multiple-choice questions adapted from 14 classic computer vision tasks.
Updated Sep 5, 2026
Models15
Model coverage15
MetricScore
EvidenceB
BLINK Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelBY | Score81.40% | Percentile100.00% | Participants15 | EvidenceC | Evaluated |
| Rank02 | ModelBY | Score79.40% | Percentile92.86% | Participants15 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score70.70% | Percentile85.71% | Participants15 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score69.10% | Percentile78.57% | Participants15 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score68.70% | Percentile71.43% | Participants15 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score68.50% | Percentile64.29% | Participants15 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score67.70% | Percentile57.14% | Participants15 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score67.30% | Percentile50.00% | Participants15 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score67.10% | Percentile42.86% | Participants15 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score65.80% | Percentile35.71% | Participants15 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score65.40% | Percentile28.57% | Participants15 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score63.40% | Percentile21.43% | Participants15 | EvidenceC | Evaluated |
| Rank13 | ModelLA | Score61.50% | Percentile14.29% | Participants15 | EvidenceC | Evaluated |
| Rank14 | ModelMI | Score61.30% | Percentile7.14% | Participants15 | EvidenceC | Evaluated |
| Rank15 | ModelCO | Score52.70% | Percentile0.00% | Participants15 | EvidenceC | Evaluated |
BLINK Highlights
The leading models and scores on this benchmark.
BLINK Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for BLINK
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is BLINK?
What BLINK measures and how its scores work.
BLINK is a benchmark for multimodal language models, with questions paired with single or multiple images and visual prompting.
It measures performance on relative depth estimation, visual correspondence, forensics detection, multi-view reasoning, counting, object localization, and spatial reasoning.
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
- No
- Evaluation key
- overall
LLM Stats. 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 81.40%.
What are the top three models on BLINK?
The current leaders are Seed 2.1 Pro (81.40%), Seed 2.1 Turbo (79.40%), and Qwen3 VL 235B A22B Instruct (70.70%).
Which BLINK model has the lowest official input price?
No matched official input price is currently available.
Which models are fastest among BLINK results?
The fastest matched records are Phi-4-multimodal-instruct (25.00 tok/s via DeepInfra) and Qwen3 VL 4B Thinking (8.77 tok/s via DeepInfra).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.