MuirBench

AI benchmark results and model performance

Type: percentageMultimodal

Recorded Results

Average Score

71.26%

Score Range

59.20% - 81.80%

Leading Model (lowest score)

59.20% - Qwen 2.5 Coder 3B

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 Coder 3B	-	59.20%	inferred family alias from qwen2.5-omni-7b (score=0.3000; benches=45)	Yes	Source
Qwen	Qwen 2.5 Coder 7B	-	59.20%	inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4700; benches=45)	Yes	Source
Qwen	Qwen 2.5 Math 7B	-	59.20%	inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4767; benches=45)	Yes	Source
Qwen	Qwen 2.5 Math 7B PRM800K	-	59.20%	inferred family alias from qwen2.5-omni-7b (score=0.3696; benches=45)	Yes	Source
Qwen	Qwen 2.5 Math PRM 7B	-	59.20%	inferred family alias from qwen2.5-omni-7b (score=0.4092; benches=45)	Yes	Source
Qwen	Qwen 2.5 Omni 3B	-	59.20%	inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4933; benches=45)	Yes	Source
Qwen	Qwen 2.5 Omni 7B	-	59.20%	-	Yes	Source
Qwen	Qwen 3 VL 30B A3B Instruct	-	62.90%	-	Yes	-
Qwen	Qwen 3 VL 4B Instruct	-	63.80%	-	Yes	-
Qwen	Qwen 3 VL 8B Instruct	-	64.40%	-	Yes	-
Qwen	Qwen 3 VL 235B A22B Instruct	-	72.80%	-	Yes	-
Qwen	Qwen 3 VL 32B Instruct	-	72.80%	-	Yes	-
Qwen	Qwen 3 4B Instruct 2507	-	75%	inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48)	Yes	-
Qwen	Qwen 3 4B SafeRL	-	75%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 4B Thinking 2507	-	75%	inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48)	Yes	-
Qwen	Qwen 3 Embedding 4B	-	75%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 Guard Gen 4B	-	75%	inferred family alias from qwen3-vl-4b-thinking (score=0.3400; benches=48)	Yes	-
Qwen	Qwen 3 Guard Stream 4B	-	75%	inferred family alias from qwen3-vl-4b-thinking (score=0.3371; benches=48)	Yes	-
Qwen	Qwen 3 Reranker 4B	-	75%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 VL 4B Thinking	-	75%	-	Yes	-
ByteDance	Seed 2.0 Lite	14 Feb 2026	76.20%	Seed2 official benchmark table \| MUIRBench	Yes	Source
Qwen	Qwen 3 8B	-	76.80%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50)	Yes	-
Qwen	Qwen 3 Guard Stream 8B	-	76.80%	inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50)	Yes	-
Qwen	Qwen 3 VL 8B Thinking	-	76.80%	-	Yes	-
Qwen	Qwen 3 VL Embedding 8B	-	76.80%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5232; benches=50)	Yes	-
Qwen	Qwen 3 VL Reranker 8B	-	76.80%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5275; benches=50)	Yes	-
Qwen	Qwen 3 VL 30B A3B Thinking	-	77.60%	-	Yes	-
ByteDance	Seed 2.0 Mini	14 Feb 2026	78%	Seed2 official benchmark table \| MUIRBench	Yes	Source
Qwen	Qwen 3 VL 235B A22B Thinking	-	80.10%	-	Yes	-
Qwen	Qwen 3 VL 32B Thinking	-	80.30%	-	Yes	-
ByteDance	Seed 2.0 Pro	14 Feb 2026	81.80%	Seed2 official benchmark table \| MUIRBench	Yes	Source