Skip to content

Benchmark

IFEval

Instruction following benchmarks measure how closely models follow instructions. Scores are percentages. Higher is better.

Models measured
209
Scores recorded
219
Sources
3
IQ weight
0.84

Score distribution

57.9%
Mean
59.7%
Median
26.2%
Std dev
20.6%
10th percentile
89.8%
90th percentile
0%
Min
0%98.9%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.84 for chat models, measured over 209 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101Granite 3.1 2B Instruct· Hugging Face Open LLM Leaderboard data62.9%
102Pixtral 12B2409· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim61.3%
103Qwen2.5 Coder 7B Instruct· Hugging Face Open LLM Leaderboard data61%
104Yi 1.5 34B Chat· Hugging Face Open LLM Leaderboard data60.7%
105Phi-3 Medium 128K Instruct· Hugging Face Open LLM Leaderboard data60.4%
106L3.3 Nevoria R1 70b· Hugging Face Open LLM Leaderboard data60.2%
107Qwen2-VL (72B) Instruct· Hugging Face Open LLM Leaderboard data59.8%
108Phi-3 Mini 128K Instruct· Hugging Face Open LLM Leaderboard data59.8%
109OpenChat 3.5· Hugging Face Open LLM Leaderboard data59.7%
110Qwen1.5-110B-Chat· Hugging Face Open LLM Leaderboard data59.4%
111Nous Hermes 2 Mixtral 8x7B DPO· Hugging Face Open LLM Leaderboard data59%
112Nous-Hermes-2-Mixtral-8x7B-SFT· Hugging Face Open LLM Leaderboard data57.3%
113Llama 3.2 1B Instruct· Hugging Face Open LLM Leaderboard data57%
114Qwen2 7B Instruct· Hugging Face Open LLM Leaderboard data56.8%
115Gemma 2 2B IT· Hugging Face Open LLM Leaderboard data56.7%
116Phi-3 Mini 4K Instruct· Hugging Face Open LLM Leaderboard data56.1%
117Mixtral 8x7B Instruct v0.1· Hugging Face Open LLM Leaderboard data56%
118DeepSeek LLM 67B Chat· Hugging Face Open LLM Leaderboard data55.9%
119OpenHermes 2.5 Mistral 7B· Hugging Face Open LLM Leaderboard data55.7%
120arcee-blitz· Hugging Face Open LLM Leaderboard data55.4%
121internlm2-5-7b-chat· Hugging Face Open LLM Leaderboard data55.4%
122Qwen1.5-32B-Chat· Hugging Face Open LLM Leaderboard data55.3%
123Mistral 7B Instruct v0.2· Hugging Face Open LLM Leaderboard data55%
124Mistral (7B) Instruct v0.3· Hugging Face Open LLM Leaderboard data54.7%
125DBRX-instruct· Hugging Face Open LLM Leaderboard data54.2%
126NousResearch: Hermes 2 Pro - Llama-3 8B· Hugging Face Open LLM Leaderboard data53.6%
127Granite 3.0 8B Instruct· Hugging Face Open LLM Leaderboard data53.1%
128OpenHermes-2-Mistral-7B· Hugging Face Open LLM Leaderboard data52.9%
129WizardLM 2 8x22B· Hugging Face Open LLM Leaderboard data52.7%
130Granite 3.0 2B Instruct· Hugging Face Open LLM Leaderboard data51.4%
131fimbulvetr-11b-v2· Hugging Face Open LLM Leaderboard data51%
132Gemma 1.1 7B IT· Hugging Face Open LLM Leaderboard data50.4%
133Mistral-7B-OpenOrca· Hugging Face Open LLM Leaderboard data49.8%
134Llama-2-70b-chat-hf· Hugging Face Open LLM Leaderboard data49.6%
135Zephyr 7B Beta· Hugging Face Open LLM Leaderboard data49.5%
136Qwen1.5-14B-Chat· Hugging Face Open LLM Leaderboard data47.7%
137SOLAR 10.7B Instruct v1.0· Hugging Face Open LLM Leaderboard data47.4%
138Yi 34B Chat· Hugging Face Open LLM Leaderboard data47%
139MN 12b Mag Mell R1· Hugging Face Open LLM Leaderboard data46.1%
140Qwen2 VL 7b Instruct· Hugging Face Open LLM Leaderboard data46%
141Yi 1.5 34B Chat16k· Hugging Face Open LLM Leaderboard data45.6%
142DeepHermes 3 Mistral 24B Preview· Hugging Face Open LLM Leaderboard data45.4%
143Mistral (7B) Instruct v0.1· Hugging Face Open LLM Leaderboard data44.9%
144Qwen2.5 1.5B Instruct· Hugging Face Open LLM Leaderboard data44.8%
145Qwen3.5 0.8B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim44%
146DeepSeek R1 Distill Qwen 14B· Hugging Face Open LLM Leaderboard data43.8%
147Qwen1.5-7B-Chat· Hugging Face Open LLM Leaderboard data43.7%
148Qwen 2.5 Coder 32B· Hugging Face Open LLM Leaderboard data43.6%
149DeepSeek R1 Distill Llama 70B· Hugging Face Open LLM Leaderboard data43.4%
150granite-3.0-3b-a800m-instruct· Hugging Face Open LLM Leaderboard data43%

Showing 101–150 of 219 results.

Data credits: Data by LLM Stats · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.