Skip to content

Benchmark

Arena ELO - Industry Mathematical

People compare model answers and choose which they prefer. Higher scores are better.

Models measured
348
Scores recorded
402
Sources
1
IQ weight
1.25

Score distribution

1,321.4
Mean
1,367.7
Median
154.1
Std dev
1,094.5
10th percentile
1,481.4
90th percentile
833.6
Min
833.61,566.4
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.25 for chat models, measured over 348 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
201Qwen2.5-Max· LMArena data1,367.7
202Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,367.5
203Qwen3 Coder· LMArena data1,367.2
204Ring Flash 2.0· LMArena data1,366.9
205O1 Mini· LMArena data1,366.5
206INTELLECT-3· LMArena data1,366.1
207DeepSeek V30324· LMArena data1,364.1
208Qwen3 30B A3B· LMArena data1,363.7
209MiniMax M1· LMArena data1,363.1
210Hunyuan TurboS20250416· LMArena data1,361.4
211Grok 3 Minibeta· LMArena data1,360.2
212GPT 4.12025-04-14· LMArena data1,359.9
213Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data1,359.3
214Claude Sonnet 420250514· LMArena data1,358.1
215Gemini 2.0 Flash001· LMArena data1,357.2
216Kimi K2 Preview0711· LMArena data1,356.8
217QwQ 32B· LMArena data1,356.3
218Mistral Medium 32505· LMArena data1,354.3
219Trinity Large Preview· LMArena data1,353.9
220Mistral Small 3.22506· LMArena data1,352.5
221Trinity Large Thinking· LMArena data1,350.8
222Olmo 3.1 32B Instruct· LMArena data1,349.4
223GPT-4.1 mini2025-04-14· LMArena data1,346.8
224Step-1o Turbo202506· LMArena data1,344.7
225Llama 3.3 Nemotron Super 49B v1· LMArena data1,344.3
226Claude Sonnet 3.7Thinking 32K, 20250219· LMArena data1,338.9
227Gemma 3 12B IT· LMArena data1,336.1
228GLM 4.5V· LMArena data1,335.4
229Qwen Plus0125· LMArena data1,332.6
230Olmo 3 32B Think· LMArena data1,329.4
231Gemma 3 27B IT· LMArena data1,326.8
232GPT-5 nanoHigh· LMArena data1,326.1
233Granite 4.1 8B· LMArena data1,324.9
234Gemini 1.5 Pro002· LMArena data1,324.7
235Gemini 2.0 Flash-Lite Preview02-05· LMArena data1,318
236DeepSeek V3· LMArena data1,314
237Claude Sonnet 3.720250219· LMArena data1,313.8
238Command A03-2025· LMArena data1,312.6
239Olmo 3.1 32B Think· LMArena data1,310.6
240Step-2 16K Exp202412· LMArena data1,310.3
241Claude 3.5 Sonnet20241022· LMArena data1,306.8
242Yi Lightning· LMArena data1,306.6
243GPT OSS 20b· LMArena data1,306
244Athene V2 Chat· LMArena data1,305.4
245Qwen2.5-Plus1127· LMArena data1,305.1
246Hunyuan Large2025-02-10· LMArena data1,304.6
247Hunyuan Large Vision· LMArena data1,304
248Hunyuan Turbo0110· LMArena data1,298.7
249Llama 4 Maverick· LMArena data1,294.9
250Claude 3.5 Sonnet20240620· LMArena data1,293.7

Showing 201–250 of 402 results.

Data credits: LMArena, CC BY 4.0 (licence)