Benchgen

BLINK — Results

RankModelScore
1lfm2-5-vl-3b61.5
2north-micro-vision-instruct52.7
B

BLINK

1 phaseActive

Multimodal benchmark for core visual perception abilities humans perform 'in a blink' — depth, similarity, spatial reasoning, and multi-image tasks.

Overview

BLINK

Category Metric Saturation Created

Paper GitHub Dataset

Quick answer: BLINK evaluates core visual perception abilities that humans can typically solve "within a blink" — such as relative depth estimation, visual similarity, spatial reasoning, and multi-view/multi-image correspondence — but which remain challenging for multimodal LLMs. LFM2.5-VL-3B scores 61.5% as of August 2026.

At a Glance

What it tests: 14 classic computer-vision task formats (depth estimation, relative reflectance, forensic detection, jigsaw, multi-view reasoning, visual correspondence, and more) reformatted as multiple-choice visual questions.

Why it matters: Many perception tasks that are trivial for humans and classical computer-vision pipelines remain surprisingly hard for multimodal LLMs; BLINK isolates this gap rather than testing world knowledge or language reasoning.

Known limitations: Because tasks are reformatted from classical CV benchmarks into multiple-choice format, chance-level performance is non-trivial (often 25-50%), so raw scores must be interpreted relative to a random baseline.

What BLINK Measures

BLINK (Fu et al., 2024) packages 14 classic visual perception tasks — drawn from established computer vision benchmarks — into a unified multiple-choice visual question-answering format. Tasks span depth estimation, relative reflectance, forensic/manipulation detection, jigsaw puzzle solving, multi-view reasoning across several images of the same scene, and visual correspondence, among others.

The benchmark's premise is that these tasks are solvable by humans "within a blink" of an eye but require perception rather than reasoning or world knowledge — probing a different capability than benchmarks like MMMU or MathVista, which lean on knowledge and multi-step reasoning.

Benchmark Specifications

FieldValue
Task categoryReasoning (visual perception)
Metric% accuracy (multiple-choice)
SaturationLow
Created byFu et al.
Source paperFu et al. 2024
GitHubzeyofu/BLINK_Benchmark
DatasetBLINK-Benchmark/BLINK on HuggingFace

How BLINK Is Scored

Each task is posed as a multiple-choice question over one or more images; the model's score is the percentage of questions answered correctly across the 14 task types, typically reported as an overall average.

State-of-the-Art Results

RankModelScoreSourceDate
1LFM2.5-VL-3B61.5%LFM2.5-VL-3B launch blog2026-08

Score sourced from Liquid AI's LFM2.5-VL-3B announcement, August 2026.

BLINK on Benchgen

No Benchgen results yet — be the first to run BLINK.

BLINK vs Other Benchmarks

BenchmarkWhat it testsSaturation
BLINKCore visual perception across 14 classic CV task formatsLow
MuirBenchMulti-image reasoning and comparisonLow
RefCOCOReferring expression comprehension (object grounding)Medium

Run BLINK on Your Model

Benchgen lets you evaluate your own vision-language model's core perception abilities and track them across model versions.

Frequently Asked Questions

What is BLINK? BLINK is a multimodal benchmark testing core visual perception abilities — depth, similarity, spatial reasoning, multi-view correspondence — that are easy for humans but hard for multimodal LLMs.
What does a good score look like on BLINK? LFM2.5-VL-3B reports 61.5% as of August 2026; scores vary widely by model size and vision encoder quality.
Who created BLINK? BLINK was created by Fu et al.; see the original paper.
Is BLINK saturated? No — BLINK remains a low-saturation benchmark, as many of its perception tasks continue to challenge even frontier multimodal models.