MMStar

MMStar - Benchmark Leaderboard & Model Performance | AI Stats

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
Qwen	Qwen 3.6 Plus	01 Apr 2026	83.30	LLM Stats (ZeroEval)	Yes	Source
ByteDance	Seed 2.0 Pro	14 Feb 2026	83.00	Seed2 official benchmark table \| MMStar	Yes	Source
Qwen	Qwen 3.5 122B A10B	24 Feb 2026	82.90	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 3.5 35B A3B	24 Feb 2026	81.90	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 3.5 27B	24 Feb 2026	81.00	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 3.5 Flash	23 Feb 2026	81.00	LLM Stats (ZeroEval) \| inferred family alias from qwen3.5-27b (score=0.4147; benches=81)	Yes	Source
ByteDance	Seed 2.0 Lite	14 Feb 2026	80.70	Seed2 official benchmark table \| MMStar	Yes	Source
Qwen	Qwen 3 VL 32B Thinking	-	79.40	LLM Stats (ZeroEval)	Yes	Source
ByteDance	Seed 2.0 Mini	14 Feb 2026	79.10	Seed2 official benchmark table \| MMStar	Yes	Source
Qwen	Qwen 3 VL 235B A22B Thinking	-	78.70	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 3 VL 235B A22B Instruct	-	78.40	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 3 VL 32B Instruct	-	77.70	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 3 VL 30B A3B Thinking	-	75.50	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 3 8B	-	75.30	LLM Stats (ZeroEval) \| inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50)	Yes	Source
Qwen	Qwen 3 Embedding 8B	-	75.30	LLM Stats (ZeroEval) \| inferred family alias from qwen3-vl-8b-thinking (score=0.3850; benches=50)	Yes	Source
Qwen	Qwen 3 Guard Gen 8B	-	75.30	LLM Stats (ZeroEval) \| inferred family alias from qwen3-vl-8b-thinking (score=0.3400; benches=50)	Yes	Source
Qwen	Qwen 3 Guard Stream 8B	-	75.30	LLM Stats (ZeroEval) \| inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50)	Yes	Source
Qwen	Qwen 3 Reranker 8B	-	75.30	LLM Stats (ZeroEval) \| inferred family alias from qwen3-vl-8b-thinking (score=0.3850; benches=50)	Yes	Source
Qwen	Qwen 3 VL 8B Thinking	-	75.30	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 3 VL Embedding 8B	-	75.30	LLM Stats (ZeroEval) \| inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5232; benches=50)	Yes	Source
Qwen	Qwen 3 VL Reranker 8B	-	75.30	LLM Stats (ZeroEval) \| inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5275; benches=50)	Yes	Source
Qwen	Qwen 3 4B Instruct 2507	-	73.20	LLM Stats (ZeroEval) \| inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48)	Yes	Source
Qwen	Qwen 3 4B SafeRL	-	73.20	LLM Stats (ZeroEval) \| inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	Source
Qwen	Qwen 3 4B Thinking 2507	-	73.20	LLM Stats (ZeroEval) \| inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48)	Yes	Source
Qwen	Qwen 3 Embedding 4B	-	73.20	LLM Stats (ZeroEval) \| inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	Source
Qwen	Qwen 3 Guard Gen 4B	-	73.20	LLM Stats (ZeroEval) \| inferred family alias from qwen3-vl-4b-thinking (score=0.3400; benches=48)	Yes	Source
Qwen	Qwen 3 Guard Stream 4B	-	73.20	LLM Stats (ZeroEval) \| inferred family alias from qwen3-vl-4b-thinking (score=0.3371; benches=48)	Yes	Source
Qwen	Qwen 3 Reranker 4B	-	73.20	LLM Stats (ZeroEval) \| inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	Source
Qwen	Qwen 3 VL 4B Thinking	-	73.20	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 3 VL 30B A3B Instruct	-	72.10	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 3 VL 8B Instruct	-	70.90	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 2.5 Math 72B	-	70.80	LLM Stats (ZeroEval) \| inferred high-confidence family alias from qwen2.5-vl-72b (score=0.4700; benches=30)	Yes	Source
Qwen	Qwen 2.5 Math PRM 72B	-	70.80	LLM Stats (ZeroEval) \| inferred family alias from qwen2.5-vl-72b (score=0.4050; benches=30)	Yes	Source
Qwen	Qwen 2.5 Math RM 72B	-	70.80	LLM Stats (ZeroEval) \| inferred family alias from qwen2.5-vl-72b (score=0.4092; benches=30)	Yes	Source
Qwen	Qwen 2.5 VL 72B Instruct	-	70.80	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 3 VL 4B Instruct	-	69.80	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 2.5 Coder 32B Instruct	-	69.50	LLM Stats (ZeroEval) \| inferred high-confidence family alias from qwen2.5-vl-32b (score=0.4641; benches=28)	Yes	Source
Qwen	Qwen 2.5 VL 32B Instruct	-	69.50	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 2.5 VL 3B Instruct	-	69.50	LLM Stats (ZeroEval) \| inferred high-confidence family alias from qwen2.5-vl-32b (score=0.4914; benches=28)	Yes	Source
Qwen	Qwen 2.5 Coder 3B	-	64.00	LLM Stats (ZeroEval) \| inferred family alias from qwen2.5-omni-7b (score=0.3000; benches=45)	Yes	Source
Qwen	Qwen 2.5 Coder 7B	-	64.00	LLM Stats (ZeroEval) \| inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4700; benches=45)	Yes	Source
Qwen	Qwen 2.5 Math 7B	-	64.00	LLM Stats (ZeroEval) \| inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4767; benches=45)	Yes	Source
Qwen	Qwen 2.5 Math 7B PRM800K	-	64.00	LLM Stats (ZeroEval) \| inferred family alias from qwen2.5-omni-7b (score=0.3696; benches=45)	Yes	Source
Qwen	Qwen 2.5 Math PRM 7B	-	64.00	LLM Stats (ZeroEval) \| inferred family alias from qwen2.5-omni-7b (score=0.4092; benches=45)	Yes	Source
Qwen	Qwen 2.5 Omni 3B	-	64.00	LLM Stats (ZeroEval) \| inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4933; benches=45)	Yes	Source
Qwen	Qwen 2.5 Omni 7B	-	64.00	LLM Stats (ZeroEval)	Yes	Source
Qwen	Qwen 2.5 0.5B	-	63.90	LLM Stats (ZeroEval) \| inferred family alias from qwen2.5-vl-7b (score=0.3131; benches=32)	Yes	Source
Qwen	Qwen 2.5 1.5B	-	63.90	LLM Stats (ZeroEval) \| inferred family alias from qwen2.5-vl-7b (score=0.3131; benches=32)	Yes	Source
Qwen	Qwen 2.5 3B	-	63.90	LLM Stats (ZeroEval) \| inferred family alias from qwen2.5-vl-7b (score=0.3531; benches=32)	Yes	Source
Qwen	Qwen 2.5 VL 7B Instruct	-	63.90	LLM Stats (ZeroEval)	Yes	Source
DeepSeek	DeepSeek VL2	13 Dec 2024	61.30	LLM Stats (ZeroEval)	Yes	Source
DeepSeek	DeepSeek VL2 Small	13 Dec 2024	57.00	LLM Stats (ZeroEval)	Yes	Source
DeepSeek	DeepSeek VL2 Tiny	13 Dec 2024	45.90	LLM Stats (ZeroEval)	Yes	Source

Recorded Results

Average Score

Score Range

Leading Model

Models Using This Benchmark