Qwen 3 VL 2B Instruct — benchmark results

Alibaba's Apache-2.0 2B dense Qwen3-VL instruct model (October 2025), the family's smallest vision-language variant, aimed at edge devices. Provider: Alibaba. Released 2025-10-19. Access: Open.

Unified ELO 1385 ± 35, rank #1301 of 1839 rated models, from 17 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
UGI - Willingness (W/10)6.5W/10 Score62.1
TriViewBench44.07Overall (self-reported)55.6
WikiVQABench56.4Accuracy (self-reported)50
VANTAGE-Bench - Single Object Tracking30.38Single Object Tracking (%)31.2
UGI Leaderboard28.23UGI Score25.9
VANTAGE-Bench - Spatial58.28Spatial (%)12.5
VANTAGE-Bench - Video QA63.85Video QA (%)12.5
UGI - Writing15.19Writing Score6.4
VANTAGE-Bench40.25Overall (%)6.2
VANTAGE-Bench - Event Verification44.73Event Verification (%)6.2
VANTAGE-Bench - Semantic54.29Semantic (%)6.2
VANTAGE-Bench - Temporal18.03Temporal (%)6.2

Interactive version: theaggregate.ai/model?slug=qwen-3-vl-2b-instruct · How It Works · Data refreshed daily, snapshot 2026-08-05.