Skip to content

Benchmark

Arena ELO

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.19

Score distribution

1,312
Mean
1,354.8
Median
146.6
Std dev
1,084.5
10th percentile
1,462.5
90th percentile
833.5
Min
833.51,539.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.19 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
251Olmo 3.1 32B Instruct· LMArena data1,311.2
252Llama 3.3 Nemotron Super 49B v1· LMArena data1,308.4
253Gemma 3n E4B IT· LMArena data1,305.6
254Grok 22024-08-13· LMArena data1,304.6
255Yi Lightning· LMArena data1,301.6
256GPT 4o2024-05-13· LMArena data1,300.5
257Claude Sonnet 3.720250219· LMArena data1,299.2
258Olmo 3 32B Think· LMArena data1,299.2
259Qwen2.5-Plus1127· LMArena data1,299.2
260Claude 3.5 Sonnet20241022· LMArena data1,297.9
261DeepSeek V2.51210· LMArena data1,294
262Granite 4.2 3B· LMArena data1,292.8
263Athene V2 Chat· LMArena data1,291.5
264Granite 4.1 8B· LMArena data1,291.3
265Gemma 3 4B IT· LMArena data1,290.8
266GLM 4 Plus· LMArena data1,289.7
267Hunyuan Large2025-02-10· LMArena data1,288
268GPT OSS 20b· LMArena data1,287.6
269Llama 4 Maverick· LMArena data1,287.2
270Molmo 2 8B· LMArena data1,287.1
271Gemini 1.5 Flash002· LMArena data1,286.7
272GPT-4o mini2024-07-18· LMArena data1,286.4
273GPT-4.1 nano2025-04-14· LMArena data1,284.8
274Llama 3.1 405B InstructBF16· LMArena data1,283.7
275Llama 3.1 Nemotron 70B Instruct· LMArena data1,282.7
276GPT 4o2024-08-06· LMArena data1,282.6
277Qwen Max0919· LMArena data1,281.9
278Llama 3.1 405B InstructFP8· LMArena data1,281.8
279Mercury· LMArena data1,281.7
280Claude 3.5 Sonnet20240620· LMArena data1,281
281Grok 2 mini2024-08-13· LMArena data1,280.7
282Llama 4 Scout· LMArena data1,279.1
283Gemini Advanced0514· LMArena data1,278.6
284Mistral Small 3.1 24B Instruct2503· LMArena data1,277.1
285Llama 3.3 70B Instruct· LMArena data1,274.3
286Hunyuan Standard2025-02-10· LMArena data1,274
287Gemini 1.5 Pro001· LMArena data1,273.5
288GPT-4 Turbo2024-04-09· LMArena data1,271.7
289DeepSeek V2.5· LMArena data1,271.3
290Olmo 3.1 32B Think· LMArena data1,271
291Qwen2.5 72B Instruct· LMArena data1,269.1
292Mistral Large 22407· LMArena data1,266.2
293Mistral Large2411· LMArena data1,265.3
294Athene 70B0725· LMArena data1,265.2
295GPT-4 Turbo Preview1106· LMArena data1,263.5
296Hunyuan Large Vision· LMArena data1,262.9
297GPT-4 Turbo Preview0125· LMArena data1,262.2
298Claude 3 Opus20240229· LMArena data1,262
299Llama 3.1 70B Instruct· LMArena data1,260.9
300Nova Pro 1.0· LMArena data1,258.7

Showing 251–300 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)