Benchmark Almanac

Qwen-VL-Plus

Qwen-VL-Plus ranks first on 2 of the 23 benchmarks with five or more models scored.

Qwen Released January 25th, 2024 Family Qwen-VL The lab's model card Compare with others
23
Benchmarks with a score
2
Ranked first, out of five or more models
5
In the top three
Qwen Jan 2024

Qwen-VL-Plus

Closed weights Mid-size

No Epoch Capabilities Index score yet
  • Video SimpleQA 23.7% #14 /27
  • HR-Bench 49.8% #8 /8
  • SEED-Bench 66.8% #3 /9

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Knowledge and language

BenchmarkScoreRankSource
Video SimpleQA 23.7% #14 of 27 Official leaderboard
MMBench-CN v1.1 68.2% #108 of 178 Official leaderboard

Vision, audio and robotics

BenchmarkScoreRankSource
CMMMU 36.8% #1 of 9 Official leaderboard
CoreCognition 61.7% #15 of 119 Official leaderboard
MMT-Bench 62.3% #2 of 15 Official leaderboard
MME 1681.3 #3 of 17 Official leaderboard
SEED-Bench-2 71.9% #1 of 5 Official leaderboard
CCBench (MMBench Chinese Culture) 55.1% #50 of 152 Official leaderboard
SEED-Bench 66.8% #3 of 9 Official leaderboard
MM-Vet 55.7% #70 of 164 Official leaderboard
OCRBench 69.4% #7 of 11 Official leaderboard
MTVQA-Test 17.8% #12 of 18 Official leaderboard
MMMU 45.2% #103 of 136 Official leaderboard
MMBench 67% #125 of 155 Official leaderboard
MathVista 43.3% #35 of 43 Official leaderboard
MMBench-EN v1.1 64.3% #148 of 178 Official leaderboard
MMStar 39.7% #144 of 173 Official leaderboard
MathVerse 11.8% #8 of 9 Official leaderboard
CharXiv 16% #74 of 77 Official leaderboard
MATH-Vision (MathVision) 10.7% #103 of 105 Official leaderboard
CharXiv (DQ) 28.9% #66 of 67 Official leaderboard
HR-Bench 49.8% #8 of 8 Official leaderboard

Other

BenchmarkScoreRankSource
OpenVLM Leaderboard 52.3 #119 of 173 Official leaderboard