Skip to content

Benchmark

Arena ELO - Industry Life and Physical and Social Science

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

Models measured
364
Scores recorded
421
Sources
1
IQ weight
1.24

Score distribution

1,316.4
Mean
1,364.4
Median
158.9
Std dev
1,071.5
10th percentile
1,475.1
90th percentile
800.9
Min
800.91,549.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.24 for chat models, measured over 364 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
251Olmo 3.1 32B Instruct· LMArena data1,316.3
252O1 Mini· LMArena data1,315.8
253Yi Lightning· LMArena data1,315.5
254o3 Mini· LMArena data1,315.1
255Grok 22024-08-13· LMArena data1,314.4
256Olmo 3 32B Think· LMArena data1,308.9
257Granite 4.1 8B· LMArena data1,307.7
258Athene V2 Chat· LMArena data1,307.1
259Gemma 3 4B IT· LMArena data1,305.5
260Granite 4.2 3B· LMArena data1,304.2
261Mercury· LMArena data1,299.8
262Grok 2 mini2024-08-13· LMArena data1,299.6
263Gemini 1.5 Flash002· LMArena data1,299.3
264Claude Sonnet 3.7Thinking 32K, 20250219· LMArena data1,298.7
265GPT-4.1 nano2025-04-14· LMArena data1,297.3
266GPT OSS 20b· LMArena data1,296.8
267Llama 3.1 405B InstructBF16· LMArena data1,295.1
268GPT 4o2024-05-13· LMArena data1,294.9
269GLM 4 Plus· LMArena data1,294.6
270Athene 70B0725· LMArena data1,293.7
271Llama 3.1 Nemotron 70B Instruct· LMArena data1,293.1
272Hunyuan Large2025-02-10· LMArena data1,291.3
273Olmo 3.1 32B Think· LMArena data1,289.6
274DeepSeek V2.51210· LMArena data1,288.2
275Llama 4 Maverick· LMArena data1,287.7
276GPT-4o mini2024-07-18· LMArena data1,286
277Claude Sonnet 3.720250219· LMArena data1,285
278Hunyuan Standard2025-02-10· LMArena data1,284.4
279Llama 3.1 405B InstructFP8· LMArena data1,284.3
280Llama 3.3 70B Instruct· LMArena data1,284.2
281Qwen Max0919· LMArena data1,279.1
282Llama 4 Scout· LMArena data1,277.9
283Llama 3.1 70B Instruct· LMArena data1,276
284Qwen2.5 72B Instruct· LMArena data1,274.9
285GPT 4o2024-08-06· LMArena data1,274
286DeepSeek V2.5· LMArena data1,273.2
287Claude 3.5 Sonnet20240620· LMArena data1,269.9
288Claude 3.5 Sonnet20241022· LMArena data1,269.8
289Gemini Advanced0514· LMArena data1,267.8
290Gemini 1.5 Pro001· LMArena data1,267.3
291Llama 3.1 Tulu 3 70B· LMArena data1,266.8
292Mistral Large 22407· LMArena data1,266.2
293Reka Core20240904· LMArena data1,265.5
294Mistral Small 3.1 24B Instruct2503· LMArena data1,264.7
295Mistral Large2411· LMArena data1,260.3
296Hunyuan Large Vision· LMArena data1,257.5
297GPT-4 Turbo2024-04-09· LMArena data1,255.7
298Nova Pro 1.0· LMArena data1,255.2
299GPT-4 Turbo Preview1106· LMArena data1,254.9
300Claude 3 Opus20240229· LMArena data1,253.9

Showing 251–300 of 421 results.

Data credits: LMArena, CC BY 4.0 (licence)