MLVU Benchmark Leaderboard | Phaseo

MLVU

AI benchmark results and model performance

Type: percentageLong Context

Recorded Results

11

Average Score

80.03%

Score Range

70.20% - 87.30%

Leading Model (lowest score)

70.20% - Qwen 2.5 0.5B

Scores Over Time

Individual benchmark scores plotted by date.

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
Qwen	Qwen 2.5 0.5B	-	70.20%	inferred family alias from qwen2.5-vl-7b (score=0.3131; benches=32)	Yes	Source
Qwen	Qwen 2.5 1.5B	-	70.20%	inferred family alias from qwen2.5-vl-7b (score=0.3131; benches=32)	Yes	Source
Qwen	Qwen 2.5 3B	-	70.20%	inferred family alias from qwen2.5-vl-7b (score=0.3531; benches=32)	Yes	Source
Qwen	Qwen 2.5 VL 7B Instruct	-	70.20%	-	Yes	Source
Qwen	Qwen 3 VL 235B A22B Thinking	-	83.80%	-	Yes	-
Qwen	Qwen 3 VL 235B A22B Instruct	-	84.30%	-	Yes	-
Qwen	Qwen 3.5 35B A3B	24 Feb 2026	85.60%	-	Yes	Source
Qwen	Qwen 3.5 27B	24 Feb 2026	85.90%	-	Yes	Source
Qwen	Qwen 3.5 Flash	23 Feb 2026	85.90%	inferred family alias from qwen3.5-27b (score=0.4147; benches=81)	Yes	Source
Qwen	Qwen 3.6 Plus	01 Apr 2026	86.70%	-	Yes	Source
Qwen	Qwen 3.5 122B A10B	24 Feb 2026	87.30%	-	Yes	Source