What does IFEval measure?

IFEval is a knowledge benchmark in the BenchGecko catalog. 78 AI models have been tested on it. Scores range from 6.0 to 90.0 out of 100.

Which model leads on IFEval?

Llama 3.3 70B Instruct from Meta leads IFEval with a score of 90.0. The median score across 78 tested models is 40.1.

No · the top score is 90.0 out of 100 (90%). There is still meaningful room for improvement on IFEval.

Does IFEval predict performance on other benchmarks?

Yes · IFEval scores correlate 0.87 with GSM8K across 16 shared models. Models that do well on IFEval tend to do well on GSM8K.

How often is IFEval data refreshed?

BenchGecko pulls updates daily. New model scores on IFEval appear as soon as they are published by Epoch AI or the model provider.

Benchmark · KnowledgeSettled

IFEval

Name: IFEval Benchmark
Creator: BenchGecko
License: https://creativecommons.org/licenses/by/4.0/

Updated 2025-10-17

Models tested

Top score

90.0

Llama 3.3 70B Instruct

Median

40.1

min 6.0

Top-5 spread

σ 2.1

Competitive

The Frontier

Best score over time · one chart, every benchmark

Chart type

Only 21 models have been tested on IFEval · not enough history to compute a frontier yet.

Pink dots = frontier records · 1 totalClick to open model page

Full rankings

78 models tested · sorted by score

#	Model	Score	Price
1	Llama 3.3 70B Instruct· Meta	90.0	$0.10
2	Llama 3.1 70B Instruct· Meta	86.7	$0.40
3	Qwen2.5 72B Instruct· Alibaba Qwen	86.4	$0.36
4	HA Qwen2.5 72B Instruct Abliterated· HuiHui AI	85.9	n/a
5	Qwen2.5 32B Instruct· Alibaba	83.5	n/a
6	Qwen2.5 14B Instruct· Alibaba	81.4	n/a
7	Gemma 2 27B· Google DeepMind	79.8	$0.65
8	Hermes 3 70B Instruct· nousresearch	76.6	$0.70
9	Qwen2.5 7B Instruct· Alibaba Qwen	75.8	$0.10
10	Gemma 2 9B· Google DeepMind	74.4	$0.03
11	Meta Llama 3 8B Instruct· Meta	74.1	n/a
12	Llama 3.2 3B Instruct· Meta	73.9	$0.05
13	Phi 4 Mini Instruct· Microsoft	73.8	$0.08
14	Phi 4 Mini Instruct· Microsoft	73.8	n/a
15	Qwen2.5 Coder 32B Instruct· Alibaba Qwen	72.7	$0.66
16	Qwen2.5 Coder 14B Instruct· Alibaba	69.1	n/a
17	Phi 4· Microsoft	64.8	$0.07
18	Qwen2.5 3B Instruct· Alibaba	64.8	n/a
19	Qwen2.5 Coder 7B Instruct· Alibaba Qwen	61.0	$0.03
20	Qwen2.5 Coder 7B Instruct· Alibaba	61.0	n/a
21	Llama 3.2 1B Instruct· Meta	58.1	$0.03
22	Phi 3.5 Mini Instruct· Microsoft	57.8	n/a
23	Qwen2 7B Instruct· Alibaba	56.8	n/a
24	Gemma 2 2b It· Google DeepMind	56.7	n/a
25	Magnum v4 72B· anthracite-org	56.3	$3.00
26	Mistral 7B Instruct V0.2· Mistral AI	55.0	n/a
27	Phi 3 Mini 4k Instruct· Microsoft	54.8	n/a
28	Hermes 2 Pro - Llama-3 8B· nousresearch	53.6	$0.14
29	WizardLM-2 8x22B· Microsoft	52.7	$0.62
30	Llama 3.1 8B Instruct· Meta	50.6	$0.02
31	Qwen2 VL 7B Instruct· Alibaba	46.0	n/a
32	Mistral 7B Instruct v0.1· Mistral AI	44.9	$0.11
33	Qwen2.5 1.5B Instruct· Alibaba	44.8	n/a
34	DeepSeek R1 Distill Qwen 14B· DeepSeek	43.8	n/a
35	DeepSeek R1 Distill Llama 8B· DeepSeek	43.4	n/a
36	R1 Distill Llama 70B· DeepSeek	43.4	$0.80
37	DeepSeek R1 Distill Qwen 32B· DeepSeek	41.9	n/a
38	R1 Distill Qwen 32B· DeepSeek	41.9	$0.29
39	DeepSeek R1 Distill Qwen 7B· DeepSeek	40.4	n/a
40	Llama 2 7b Chat Hf· Meta	39.9	n/a
41	QwQ 32B· Alibaba Qwen	39.8	$0.15
42	D Dolphin 2.9.1 Yi 1.5 34b· DPHN	38.5	n/a
43	Qwen2-72B· Alibaba Qwen	38.2	n/a
44	U Stable Beluga 2· Unknown	37.9	n/a
45	DeepSeek R1 Distill Qwen 1.5B· DeepSeek	34.6	n/a
46	Qwen2 1.5B Instruct· Alibaba	33.7	n/a
47	Llama 3-70B· Meta	33.2	n/a
48	Falcon-180B· TII	32.6	n/a
49	Qwen2.5 0.5B Instruct· Alibaba	31.5	n/a
50	U Yi 6B· Unknown	30.5	n/a
51	HF SmolLM2 135M Instruct· Hugging Face TB	28.8	n/a
52	U StarCoder 2 15B· Unknown	27.8	n/a
53	Phi 2· Microsoft	27.4	n/a
54	Gemma 2B· Google DeepMind	26.6	n/a
55	LLaMA-13B· Meta	25.3	n/a
56	Llama 2 7b Hf· Meta	25.2	n/a
57	Llama 3 8B Instruct· Meta	24.0	$0.14
58	Mistral 7B V0.1· Mistral AI	23.9	n/a
59	L Vicuna 7b V1.5· LMSYS	23.5	n/a
60	Qwen2 0.5B Instruct· Alibaba	22.5	n/a
61	Gpt2 Medium· OpenAI	22.1	n/a
62	U MPT-30B· Unknown	21.5	n/a
63	U RedPajama-INCITE-7B-Base· Unknown	20.8	n/a
64	Gpt2 Large· OpenAI	20.5	n/a
65	Phi-1.5· Microsoft	20.3	n/a
66	Gemma 2 2b· Google DeepMind	20.2	n/a
67	Gpt2· OpenAI	19.3	n/a
68	Gpt Neo 125m· eleutherai	19.1	n/a
69	Qwen2 0.5B· Alibaba	18.7	n/a
70	HF SmolLM2 135M· Hugging Face TB	18.2	n/a
71	Pythia 160m· eleutherai	18.2	n/a
72	Llama 3.1 405B· Meta	18.1	n/a
73	U INTELLECT-1· Unknown	17.6	n/a
74	Meta Llama 3 8B· Meta	16.0	n/a
75	GLM 4 32B · z-ai	14.3	$0.10
76	Llama 3.2 3B Instruct (free)· Meta	13.4	$0.00
77	D Distilgpt2· DistilBERT	6.1	n/a
78	T TinyLlama 1.1B Chat V1.0· TinyLlama	6.0	n/a