Multi-IF

AI benchmark results and model performance

Type: percentageCommunication

Recorded Results

Average Score

71.27%

Score Range

60.90% - 80.60%

Leading Model (lowest score)

60.90% - GPT 4o (2024-08-06)

Scores Over Time

Individual benchmark scores plotted by date.

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
OpenAI	GPT 4o (2024-08-06)	06 Aug 2024	60.90%	-	Yes	Source
OpenAI	GPT 4o Audio Preview (2025-06-03)	03 Jun 2025	60.90%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Audio Preview (2024-10-01)	01 Oct 2024	60.90%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Audio Preview (2024-12-17)	17 Dec 2024	60.90%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Realtime Preview (2025-06-03)	03 Jun 2025	60.90%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Realtime Preview (2024-10-01)	01 Oct 2024	60.90%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Search Preview	11 Mar 2025	60.90%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Transcribe	20 Mar 2025	60.90%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Transcribe Diarize	15 Oct 2025	60.90%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
Qwen	Qwen 3 VL 30B A3B Instruct	-	66.10%	-	Yes	-
OpenAI	GPT 4.5 Preview	27 Feb 2025	70.80%	-	Yes	Source
Qwen	Qwen 3 VL 32B Instruct	-	72%	-	Yes	-
Qwen	Qwen 3 30B A3B	-	72.20%	-	Yes	Source
Qwen	Qwen 3 30B A3B Instruct 2507	-	72.20%	inferred version-family alias from qwen3-30b-a3b	Yes	Source
Qwen	Qwen 3 30B A3B Thinking 2507	-	72.20%	inferred version-family alias from qwen3-30b-a3b	Yes	Source
Qwen	Qwen 3 Coder 30B A3B Instruct	-	72.20%	inferred high-confidence family alias from qwen3-30b-a3b (score=0.5007; benches=8)	Yes	Source
Qwen	Qwen 3 Omni 30B A3B Captioner	-	72.20%	inferred family alias from qwen3-30b-a3b (score=0.4129; benches=8)	Yes	Source
Qwen	Qwen 3 Omni 30B A3B Instruct	-	72.20%	inferred high-confidence family alias from qwen3-30b-a3b (score=0.4819; benches=8)	Yes	Source
Qwen	Qwen 3 Omni 30B A3B Thinking	-	72.20%	inferred high-confidence family alias from qwen3-30b-a3b (score=0.4819; benches=8)	Yes	Source
Qwen	Qwen 3 VL 30B A3B Thinking	-	73%	-	Yes	-
Qwen	Qwen 3 4B Instruct 2507	-	73.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48)	Yes	-
Qwen	Qwen 3 4B SafeRL	-	73.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 4B Thinking 2507	-	73.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48)	Yes	-
Qwen	Qwen 3 Embedding 4B	-	73.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 Guard Gen 4B	-	73.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3400; benches=48)	Yes	-
Qwen	Qwen 3 Guard Stream 4B	-	73.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3371; benches=48)	Yes	-
Qwen	Qwen 3 Reranker 4B	-	73.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 VL 4B Thinking	-	73.60%	-	Yes	-
Qwen	Qwen 3 8B	-	75.10%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50)	Yes	-
Qwen	Qwen 3 Guard Stream 8B	-	75.10%	inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50)	Yes	-
Qwen	Qwen 3 VL 8B Instruct	-	75.10%	-	Yes	-
Qwen	Qwen 3 VL 8B Thinking	-	75.10%	-	Yes	-
Qwen	Qwen 3 VL Embedding 8B	-	75.10%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5232; benches=50)	Yes	-
Qwen	Qwen 3 VL Reranker 8B	-	75.10%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5275; benches=50)	Yes	-
Qwen	Qwen 3 Next 80B A3B Instruct	-	75.80%	MultiIF	Yes	Source
Qwen	Qwen 3 VL 235B A22B Instruct	-	76.30%	-	Yes	-
Qwen	Qwen 3 Next 80B A3B Thinking	-	77.80%	MultiIF	Yes	Source
Qwen	Qwen 3 VL 32B Thinking	-	78%	-	Yes	-
Qwen	Qwen 3 VL 235B A22B Thinking	-	79.10%	-	Yes	-
OpenAI	o3 mini	30 Jan 2025	79.50%	-	Yes	Source
Qwen	Qwen 3 235B A22B Thinking 2507	-	80.60%	MultiIF	Yes	Source