Skip to content

Benchmark

Arena ELO - Industry Entertainment and Sports and Media

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

Models measured
364
Scores recorded
421
Sources
1
IQ weight
1.17

Score distribution

1,279.8
Mean
1,314.7
Median
149.5
Std dev
1,058.5
10th percentile
1,441.1
90th percentile
784.8
Min
784.81,526.7
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.17 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
251Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data1,270.4
252Qwen Plus0125· LMArena data1,270.1
253Olmo 3.1 32B Instruct· LMArena data1,269.9
254GPT-5 nanoHigh· LMArena data1,268.8
255Granite 4.2 8B· LMArena data1,268.2
256GPT 4o2024-08-06· LMArena data1,267.5
257Hunyuan Turbo0110· LMArena data1,266.7
258Olmo 3 32B Think· LMArena data1,263.7
259Gemma 3n E4B IT· LMArena data1,262.3
260Hunyuan Large2025-02-10· LMArena data1,262
261GPT-4 Turbo2024-04-09· LMArena data1,258.2
262Qwen3 30B A3B· LMArena data1,258.2
263O1 Mini· LMArena data1,256.9
264GPT-4o mini2024-07-18· LMArena data1,253.3
265Hunyuan TurboS20250226· LMArena data1,252.9
266Gemini 1.5 Flash002· LMArena data1,251.5
267Gemini 1.5 Pro001· LMArena data1,251.5
268Mistral Small 3.1 24B Instruct2503· LMArena data1,250.2
269Llama 3.1 405B InstructFP8· LMArena data1,250.1
270GLM 4 Plus· LMArena data1,249.9
271Llama 4 Maverick· LMArena data1,249.7
272Llama 3.1 Nemotron 70B Instruct· LMArena data1,248.9
273Granite 4.1 8B· LMArena data1,245
274Qwen2.5-Plus1127· LMArena data1,244.3
275GPT-4.1 nano2025-04-14· LMArena data1,242.7
276GPT-4 Turbo Preview1106· LMArena data1,242.6
277Llama 4 Scout· LMArena data1,241.9
278Mistral Large 22407· LMArena data1,241.6
279Llama 3.1 405B InstructBF16· LMArena data1,241
280Llama 3.3 70B Instruct· LMArena data1,240.7
281Qwen Max0919· LMArena data1,240.6
282Magistral Medium2506· LMArena data1,240.5
283Llama 3.1 Tulu 3 70B· LMArena data1,240.1
284Claude 3.5 Sonnet20240620· LMArena data1,238.6
285GPT-4 Turbo Preview0125· LMArena data1,237
286Olmo 3.1 32B Think· LMArena data1,236.7
287Grok 2 mini2024-08-13· LMArena data1,232.9
288Gemma 3 4B IT· LMArena data1,232.8
289Athene 70B0725· LMArena data1,232.6
290Mistral Large2411· LMArena data1,230.5
291DeepSeek V2.5· LMArena data1,230
292Llama 3.1 70B Instruct· LMArena data1,229.9
293Athene V2 Chat· LMArena data1,228.8
294OLMo 2 32B Instruct0325· LMArena data1,224.5
295Claude Haiku 3.520241022· LMArena data1,224
296Command R+08-2024· LMArena data1,223.4
297Hunyuan Standard2025-02-10· LMArena data1,222.4
298Claude 3 Opus20240229· LMArena data1,220.1
299Hunyuan Large Vision· LMArena data1,219.1
300Jamba 1.5 Large· LMArena data1,218.2

Showing 251–300 of 421 results.

Data credits: LMArena, CC BY 4.0 (licence)