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
51GPT-4.1 mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim84.1%
52Mistral Large Instruct 2411· Hugging Face Open LLM Leaderboard data84%
53QwQ 32B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
83.9%
54Qwen3 VL 8B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.83.7%
55Qwen2.5 32B Instruct· Hugging Face Open LLM Leaderboard data83.5%
56Phi-4-reasoning· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim83.4%
57Qwen: Qwen3 VL 8B Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.83.2%
58Mistral Small 24B Instruct2501· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim82.9%
59qwen3-vl-30bthinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.81.7%
60Qwen 2.5 14B Instruct· Hugging Face Open LLM Leaderboard data81.6%
61GPT 4o· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim81%
62Llama 3 70B Instruct· Hugging Face Open LLM Leaderboard data81%
63Llama 3.1 8B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
80.4%
64Gemma 3 1b it· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.2%
65Qwen2 72B Instruct· Hugging Face Open LLM Leaderboard data79.9%
66Gemma 2 27B IT· Hugging Face Open LLM Leaderboard data79.8%
67Qwen3.5 2B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim78.6%
68Qwen2 72B Dracarys· Hugging Face Open LLM Leaderboard data78.6%
69Llama 3 70b Instruct Abliterated v3.5· Hugging Face Open LLM Leaderboard data77.5%
70Llama 3.2 3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
77.4%
71C4AI Command R7b 12 2024· Hugging Face Open LLM Leaderboard data77.1%
72Hermes 3 Llama 3.1 70B· Hugging Face Open LLM Leaderboard data76.6%
73Qwen2.5 7B Instruct· Hugging Face Open LLM Leaderboard data
Values from 2 sources
75.9%
74Neural Daredevil 8B abliterated· Hugging Face Open LLM Leaderboard data75.6%
75Magnum V2 72B· Hugging Face Open LLM Leaderboard data75.6%
76GPT-4.1 nano· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim74.5%
77Gemma 2 9B IT· Hugging Face Open LLM Leaderboard data74.4%
78Llama 3 8B Instruct· Hugging Face Open LLM Leaderboard data74.1%
79l3-70b-euryale-v2.1· Hugging Face Open LLM Leaderboard data73.8%
80Llama 3.1 Nemotron 70B Instruct HF· Hugging Face Open LLM Leaderboard data73.8%
81Phi 4 Mini· Hugging Face Open LLM Leaderboard data73.8%
82Aya Expanse 32B· Hugging Face Open LLM Leaderboard data73%
83Granite 3.2 8b Instruct· Hugging Face Open LLM Leaderboard data72.7%
84Qwen 2.5 Coder 32B Instruct· Hugging Face Open LLM Leaderboard data72.7%
85Granite 3.1 8B Instruct· Hugging Face Open LLM Leaderboard data72.1%
86Mixtral 8x22B Instruct v0.1· Hugging Face Open LLM Leaderboard data71.8%
87L3.3 Ms Nevoria 70b· Hugging Face Open LLM Leaderboard data69.6%
88phi-3.5-moe-instruct· Hugging Face Open LLM Leaderboard data69.2%
89qwen2.5-coder-14b-instruct· Hugging Face Open LLM Leaderboard data69.1%
90L3 8B Lunaris V1v1· Hugging Face Open LLM Leaderboard data68.9%
91EVA-Qwen2.5-72B-v0.2· Hugging Face Open LLM Leaderboard data68.8%
92L3 8B Stheno V3.2· Hugging Face Open LLM Leaderboard data68.7%
93Phi-3 Small 8K Instruct· Hugging Face Open LLM Leaderboard data65%
94Qwen2.5 3B Instruct· Hugging Face Open LLM Leaderboard data64.7%
95Phi-3 Medium 4K Instruct· Hugging Face Open LLM Leaderboard data64.2%
96Mistral Nemo· Hugging Face Open LLM Leaderboard data63.8%
97Aya Expanse 8B· Hugging Face Open LLM Leaderboard data63.6%
98dolphin-2-9-2-qwen2-72b· Hugging Face Open LLM Leaderboard data63.4%
99Qwen2 57B A14B Instruct· Hugging Face Open LLM Leaderboard data63.4%
100Phi 4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
63%

Showing 51–100 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.