Recent studyBlind TTS Elo is live. Compare two anonymous voice samples, vote after listening, and help separate real preference signal from noise.Vote in the study ->
Codesota · Models · Qwen2-VL-2B10 results · 10 benchmarks
Model card

Qwen2-VL-2B.

unknown
§ 02 · Benchmarks

Every benchmark Qwen2-VL-2B has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01TextVQAMultimodal · Visual Question Answeringaccuracy79.7%#15/23source ↗
02DocVQAComputer Vision · Document Understandinganls90.1%#17/21source ↗
03MMBenchMultimodal · Visual Question Answeringaccuracy74.9%#18/20source ↗
04MVBenchMultimodal · Video Understandingaccuracy63.2%#20/20source ↗
05MMStarMultimodal · Image-Text-to-Textaccuracy48.0%#21/21source ↗
06RealWorldQAMultimodal · Visual Question Answeringaccuracy62.9%#23/23source ↗
07Video-MMEMultimodal · Video Understandingaccuracy55.6%#24/24source ↗
08MMMU-ProMultimodal · Visual Question Answeringaccuracy37.6%#28/31source ↗
09MMMUMultimodal · Visual Question Answeringaccuracy41.1%#30/30source ↗
10MMMUMultimodal · Image-Text-to-Textaccuracy41.1%#36/36source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area

Where Qwen2-VL-2B actually performs.

Computer Vision
1
benchmark
avg rank #17.0
Multimodal
9
benchmarks
avg rank #23.9
§ 04 · Papers

1 paper with results for Qwen2-VL-2B.

  1. 2024-09-18· 10 results

    Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution

§ 06 · Sources & freshness

Where these numbers come from.

pwc-dump
10
results
0 of 10 rows marked verified.