IFEval Benchmark

IFEval Benchmark | Phaseo

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
Qwen	Qwen 3.5 27B	24 Feb 2026	95%	-	Yes	Source
Qwen	Qwen 3.5 Flash	23 Feb 2026	95%	inferred family alias from qwen3.5-27b (score=0.4147; benches=81)	Yes	Source
Qwen	Qwen 3.6 Plus	01 Apr 2026	94.30%	-	Yes	Source
OpenAI	o3 mini	30 Jan 2025	93.90%	-	Yes	Source
Qwen	Qwen 3.5 122B A10B	24 Feb 2026	93.40%	-	Yes	Source
Qwen	Qwen 3.5 397B A17B	16 Feb 2026	92.60%	-	Yes	Source
Meta	Llama 2 70B Chat	20 Jun 2023	92.10%	inferred family alias from llama-3.3-70b-instruct (score=0.3129; benches=9)	Yes	Source
Qwen	Qwen 3.5 35B A3B	24 Feb 2026	91.90%	-	Yes	Source
Qwen	Qwen 3.5 9B	02 Mar 2026	91.50%	-	Yes	Source
Nvidia	Nvidia Nemotron Nano 12B V2	-	90.30%	inferred high-confidence family alias from nvidia-nemotron-nano-9b-v2 (score=0.4889; benches=6)	Yes	Source
Nvidia	Nvidia Nemotron Nano 9B V2	-	90.30%	-	Yes	Source
Moonshot	Kimi K2 (2025-09-05)	05 Sept 2025	89.80%	Prompt Strict	Yes	Source
Qwen	Qwen 3.5 4B	02 Mar 2026	89.80%	-	Yes	Source
IBM	Granite 4.1 30B	29 Apr 2026	89.65%	Avg	Yes	Source
Meituan	LongCat Flash Chat	-	89.65%	inferred high-confidence family alias from longcat-flash-chat (score=0.4667; benches=16)	Yes	Source
Nvidia	Llama 3.1 Nemotron Ultra 253B v1	07 Apr 2025	89.45%	-	Yes	Source
Qwen	Qwen 3 Next 80B A3B Thinking	-	88.90%	-	Yes	Source
Qwen	Qwen 3 A235 A22B Instruct 2507	-	88.70%	-	Yes	Source
OpenAI	GPT 4.5	27 Feb 2025	88.20%	-	Yes	Source
Qwen	Qwen 3 VL 235B A22B Thinking	-	88.20%	-	Yes	-
Qwen	Qwen 3 235B A22B Thinking 2507	-	87.80%	-	Yes	Source
Qwen	Qwen 3 VL 235B A22B Instruct	-	87.80%	-	Yes	-
Qwen	Qwen 3 VL 32B Thinking	-	87.80%	-	Yes	-
Qwen	Qwen 3 Next 80B A3B Instruct	-	87.60%	-	Yes	Source
Moonshot	Kimi K1.5	20 Jan 2025	87.20%	-	Yes	Source
IBM	Granite 4.1 8B	29 Apr 2026	87.06%	Avg	Yes	Source
DeepSeek	DeepSeek OCR	20 Oct 2025	86.10%	inferred family alias from deepseek-v3 (score=0.3000; benches=20)	Yes	Source
DeepSeek	DeepSeek V2 (2024-06-28)	28 Jun 2024	86.10%	inferred family alias from deepseek-v3 (score=0.4159; benches=20)	Yes	Source
DeepSeek	DeepSeek V4	-	86.10%	inferred high-confidence family alias from deepseek-v3 (score=0.5818; benches=20)	Yes	Source
Qwen	Qwen 3 VL 30B A3B Instruct	-	85.80%	-	Yes	-
Microsoft	Phi 4 Reasoning Plus	30 Apr 2025	84.90%	-	Yes	Source
LG	EXAONE 4.0 32B	15 Jul 2025	84.80%	Non Reasoning	Yes	Source
Qwen	Qwen 3 VL 32B Instruct	-	84.70%	-	Yes	-
Qwen	Qwen 72B	-	84.10%	inferred family alias from qwen-2.5-72b-instruct (score=0.3060; benches=14)	Yes	Source
ai21	Jamba Large 1.7	03 Jul 2025	84%	-	Yes	-
Qwen	QwQ 32B	-	83.90%	-	Yes	Source
Qwen	Qwen 3 VL 8B Instruct	-	83.70%	-	Yes	-
Microsoft	Phi 4 Reasoning	30 Apr 2025	83.40%	-	Yes	Source
Qwen	Qwen 3 8B	-	83.20%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50)	Yes	-
Qwen	Qwen 3 Embedding 8B	-	83.20%	inferred family alias from qwen3-vl-8b-thinking (score=0.3850; benches=50)	Yes	-
Qwen	Qwen 3 Guard Gen 8B	-	83.20%	inferred family alias from qwen3-vl-8b-thinking (score=0.3400; benches=50)	Yes	-
Qwen	Qwen 3 Guard Stream 8B	-	83.20%	inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50)	Yes	-
Qwen	Qwen 3 Reranker 8B	-	83.20%	inferred family alias from qwen3-vl-8b-thinking (score=0.3850; benches=50)	Yes	-
Qwen	Qwen 3 VL 8B Thinking	-	83.20%	-	Yes	-
Qwen	Qwen 3 VL Embedding 8B	-	83.20%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5232; benches=50)	Yes	-
Qwen	Qwen 3 VL Reranker 8B	-	83.20%	inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5275; benches=50)	Yes	-
Qwen	Qwen 3 4B Instruct 2507	-	82.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48)	Yes	-
Qwen	Qwen 3 4B SafeRL	-	82.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 4B Thinking 2507	-	82.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48)	Yes	-
Qwen	Qwen 3 Embedding 4B	-	82.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 Guard Gen 4B	-	82.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3400; benches=48)	Yes	-
Qwen	Qwen 3 Guard Stream 4B	-	82.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3371; benches=48)	Yes	-
Qwen	Qwen 3 Reranker 4B	-	82.60%	inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48)	Yes	-
Qwen	Qwen 3 VL 4B Thinking	-	82.60%	-	Yes	-
IBM	Granite 4.1 3B	29 Apr 2026	82.30%	Avg	Yes	Source
Qwen	Qwen 3 VL 4B Instruct	-	82.30%	-	Yes	-
Qwen	Qwen 3 VL 30B A3B Thinking	-	81.70%	-	Yes	-
OpenAI	GPT 4o (2024-08-06)	06 Aug 2024	81%	-	Yes	Source
OpenAI	GPT 4o Audio (2025-06-03)	03 Jun 2025	81%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Audio (2024-10-01)	01 Oct 2024	81%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Audio (2024-12-17)	17 Dec 2024	81%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Realtime Preview (2025-06-03)	03 Jun 2025	81%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Realtime Preview (2024-10-01)	01 Oct 2024	81%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Search Preview	11 Mar 2025	81%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Transcribe	20 Mar 2025	81%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
OpenAI	GPT 4o Transcribe Diarize	15 Oct 2025	81%	inferred modality/version alias from gpt-4o-2024-08-06	Yes	Source
Nvidia	Llama 3.1 Nemotron Nano 4B V1.1	-	79.30%	inferred high-confidence family alias from llama-3.1-nemotron-nano-8b-v1 (score=0.5523; benches=7)	Yes	Source
Nvidia	Llama 3.1 Nemotron Nano 8B V1	18 Mar 2025	79.30%	-	Yes	Source
Qwen	Qwen 3.5 2B	02 Mar 2026	78.60%	-	Yes	Source
ai21	Jamba Mini 1.7	03 Jul 2025	76.20%	-	Yes	-
ai21	Jamba Large 1.6	06 Mar 2025	75.80%	-	Yes	-
IBM	Granite 3.0 8B Instruct	-	74.82%	inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14)	Yes	Source
IBM	Granite 3.1 8B Instruct	-	74.82%	inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14)	Yes	Source
IBM	Granite 3.2 8B Instruct	-	74.82%	inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14)	Yes	Source
IBM	Granite 3.2 8B Instruct Preview	-	74.82%	inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4687; benches=14)	Yes	Source
IBM	Granite 3.3 2B Instruct	16 Apr 2025	74.82%	inferred family alias from granite-3.3-8b-instruct (score=0.3627; benches=14)	Yes	Source
IBM	Granite 3.3 8B Instruct	16 Apr 2025	74.82%	-	Yes	Source
IBM	Granite Guardian 3.0 8B	-	74.82%	inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14)	Yes	Source
IBM	Granite Guardian 3.1 8B	-	74.82%	inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14)	Yes	Source
IBM	Granite Guardian 3.3 8B	-	74.82%	inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.5071; benches=14)	Yes	Source
IBM	Granite Speech 3.2 8B	-	74.82%	inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14)	Yes	Source
IBM	Granite Speech 3.3 8B	-	74.82%	inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.5071; benches=14)	Yes	Source
LG	EXAONE 4.0 1.2B	15 Jul 2025	74.70%	Non Reasoning	Yes	Source
Qwen	Qwen 7B	-	71.20%	inferred family alias from qwen-2.5-7b-instruct (score=0.3083; benches=14)	Yes	Source
ai21	Jamba Mini 1.6	06 Mar 2025	68.30%	-	Yes	-
IBM	Granite 4.0 Micro	02 Oct 2025	63%	inferred high-confidence family alias from granite-4.0-tiny-preview (score=0.4700; benches=12)	Yes	Source
IBM	Granite 4.0 Small	02 Oct 2025	63%	inferred high-confidence family alias from granite-4.0-tiny-preview (score=0.4700; benches=12)	Yes	Source
IBM	Granite 4.0 Tiny	02 Oct 2025	63%	inferred alias from granite-4.0-tiny-preview	Yes	Source
IBM	Granite 4.0 Tiny Preview	02 May 2025	63%	-	Yes	Source
Microsoft	Phi 1	-	63%	inferred family alias from phi-4 (score=0.3100; benches=13)	Yes	Source
Microsoft	Phi 2	-	63%	inferred family alias from phi-4 (score=0.3100; benches=13)	Yes	Source
Microsoft	Phi 4	12 Dec 2024	63%	-	Yes	Source
Mistral	Pixtral 12B	17 Sept 2024	61.30%	inferred version-family alias from pixtral-12b-2409	Yes	Source
Qwen	Qwen 3.5 0.8B	02 Mar 2026	44%	-	Yes	Source

Average Score

Score Range

Leading Model

IFEval

Recorded Results

Average Score

Score Range

Leading Model

Models Using This Benchmark