Skip to content

Benchmark

Arena ELO - Instruction Following

Instruction following benchmarks measure how closely models follow instructions. Higher scores are better.

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.18

Score distribution

1,294.5
Mean
1,332.7
Median
155.5
Std dev
1,056.7
10th percentile
1,465
90th percentile
780.7
Min
780.71,539.1
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.18 for chat models, measured over 366 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
201DeepSeek V30324· LMArena data1,344.4
202Qwen3 Next 80B A3Bthinking· LMArena data1,344.1
203Hunyuan TurboS20250416· LMArena data1,343.6
204O1 Preview· LMArena data1,341.6
205Mistral Medium 32505· LMArena data1,338.3
206o3 MiniHigh· LMArena data1,337.3
207Gemini 2.0 Flash001· LMArena data1,335.7
208Nova 2 Lite· LMArena data1,335.1
209Qwen2.5-Max· LMArena data1,334.7
210Trinity Large Preview· LMArena data1,334.3
211Qwen 3 235B A22B· LMArena data1,332.7
212GPT-4.1 mini2025-04-14· LMArena data1,332.6
213Step3· LMArena data1,331.9
214Mercury 2· LMArena data1,329.4
215MiniMax M2· LMArena data1,328.8
216Trinity Large Thinking· LMArena data1,326.1
217GLM 4.7 Flash· LMArena data1,325.6
218Claude Sonnet 3.720250219· LMArena data1,324.4
219Kimi K2 Preview0711· LMArena data1,324.3
220Gemma 3 27B IT· LMArena data1,320.6
221o4 Mini2025-04-16· LMArena data1,320.2
222Ling Flash 2.0· LMArena data1,318.9
223Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data1,318.7
224GPT OSS 120B· LMArena data1,317.9
225INTELLECT-3· LMArena data1,315.9
226Hunyuan TurboS20250226· LMArena data1,315.5
227o3 Mini· LMArena data1,315.2
228DeepSeek V3· LMArena data1,314.9
229nova-experimental-chat10-09· LMArena data1,314.3
230Ring Flash 2.0· LMArena data1,313.5
231MiniMax M1· LMArena data1,311.8
232GLM 4.5V· LMArena data1,310.6
233Step-1o Turbo202506· LMArena data1,310
234Mistral Small 3.22506· LMArena data1,309.6
235Command A03-2025· LMArena data1,308.4
236Llama 3.1 Nemotron Ultra 253B v1· LMArena data1,307.6
237GPT-5 nanoHigh· LMArena data1,305.7
238Qwen3 32B· LMArena data1,304.6
239Gemini 2.0 Flash-Lite Preview02-05· LMArena data1,304.6
240O1 Mini· LMArena data1,303.9
241Nemotron 3 Nano 30B A3BBF16· LMArena data1,303.4
242Qwen Plus0125· LMArena data1,302.8
243Granite 4.2 8B· LMArena data1,300
244Gemma 3 12B IT· LMArena data1,299.1
245Llama 3.3 Nemotron Super 49B v1.5· LMArena data1,299.1
246Olmo 3.1 32B Instruct· LMArena data1,298.8
247Claude 3.5 Sonnet20241022· LMArena data1,297.1
248QwQ 32B· LMArena data1,296.9
249Gemini 1.5 Pro002· LMArena data1,296.8
250Llama 3.3 Nemotron Super 49B v1· LMArena data1,293.3

Showing 201–250 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)