Qwen 2 VL 2B — benchmark results

Provider: Alibaba. Released 2024-08-29. Access: Open.

Unified ELO 1334 ± 8, rank #1461 of 1776 rated models, from 1028 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
OpenVLM MMT-Bench - Artwork Emotion Recognition70Score (%)100
OpenVLM MMBench V1.1 EN - Social Relation93.4Accuracy (%)99.3
OpenVLM MMT-Bench - 3D CAD Recognition80Score (%)98.5
MEGA-Bench Task - MFC Bench Check Face Swap71.4Task Score (%)98.1
OpenVLM MMT-Bench - AR59Score (%)98.1
OpenVLM MMT-Bench - Counting By Visual Prompting55Score (%)98.1
OpenVLM VCR61.3Overall Jaccard (%)97.9
OpenVLM MMT-Bench - Geometrical Perspective91.7Score (%)96.6
OpenVLM MMBench V1.1 EN - Image Topic98.9Accuracy (%)95.2
Open LMM Reasoning - WeMath - RoteMemorization (Loose)22.3Accuracy (%)95.1
OpenVLM MMT-Bench - Google Apps55Score (%)95.1
OpenVLM MMT-Bench - Animals Recognition100Score (%)94.9

Interactive version: theaggregate.ai/model?slug=qwen-2-vl-2b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.