Qwen 2 VL 7B Instruct — benchmark results

Alibaba's Apache-2.0 7B vision-language model (August 2024) handling image, multi-image, and video input. Provider: Alibaba. Released 2024-08-29. Access: Open.

Unified ELO 1439 ± 13, rank #1046 of 1776 rated models, from 83 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
TempCompass67.8Avg All (%)85.7
Open LLM Leaderboard - GPQA9.28Score74.8
Open LLM Leaderboard - MMLU-Pro34.39Score74.6
Open LLM Leaderboard - MuSR13.55Score74.6
Open LLM Leaderboard - BBH35.88Score73
Open LLM Leaderboard - MATH Level 519.86Score70.7
KOFFVQA - Korean Recognition40Score (%)66.7
KOFFVQA - Document Understanding64.33Score (%)60.5
KOFFVQA - Object Attributes73.17Score (%)57.4
Open Japanese LLM - CG20.28Score (%)55.8
Open Japanese LLM - Mbpp Pylint Check40.16Score (%)54.6
KOFFVQA - Korean OCR70Score (%)53.1

Interactive version: theaggregate.ai/model?slug=qwen-2-vl-7b-instruct · How the rankings work · Data refreshed daily, snapshot 2026-07-22.