Creative Writing v3

AI benchmark results and model performance

Type: percentageCreativity

Recorded Results

Average Score

156%

Score Range

76.10% - 1722%

Leading Model (lowest score)

76.10% - Qwen 3 4B Instruct 2507

Scores Over Time

Individual benchmark scores plotted by date.

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
Qwen	Qwen 3 4B Instruct 2507	-	76.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48)	Yes	-
Qwen	Qwen 3 4B SafeRL	-	76.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 4B Thinking 2507	-	76.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48)	Yes	-
Qwen	Qwen 3 Embedding 4B	-	76.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 Guard Gen 4B	-	76.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3400; benches=48)	Yes	-
Qwen	Qwen 3 Guard Stream 4B	-	76.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3371; benches=48)	Yes	-
Qwen	Qwen 3 Reranker 4B	-	76.10%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 VL 4B Thinking	-	76.10%	-	Yes	-
Qwen	Qwen 3 8B	-	82.40%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50)	Yes	-
Qwen	Qwen 3 Guard Stream 8B	-	82.40%	inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50)	Yes	-
Qwen	Qwen 3 VL 8B Thinking	-	82.40%	-	Yes	-
Qwen	Qwen 3 VL Embedding 8B	-	82.40%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5232; benches=50)	Yes	-
Qwen	Qwen 3 VL Reranker 8B	-	82.40%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5275; benches=50)	Yes	-
Qwen	Qwen 3 VL 30B A3B Thinking	-	82.50%	-	Yes	-
Qwen	Qwen 3 VL 32B Thinking	-	83.30%	-	Yes	-
Qwen	Qwen 3 VL 30B A3B Instruct	-	84.60%	-	Yes	-
Qwen	Qwen 3 Next 80B A3B Instruct	-	85.30%	-	Yes	Source
Qwen	Qwen 3 VL 32B Instruct	-	85.60%	-	Yes	-
Qwen	Qwen 3 VL 235B A22B Thinking	-	85.70%	-	Yes	-
Qwen	Qwen 3 235B A22B Thinking 2507	-	86.10%	-	Yes	Source
Qwen	Qwen 3 VL 235B A22B Instruct	-	86.50%	-	Yes	-
SpaceXAI	Grok 4.1 Thinking	17 Nov 2025	1722%	inferred modality/version alias from grok-4.1-thinking-2025-11-17	Yes	Source