Skip to content

Benchmark

Arena ELO Style-Controlled

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.10

Score distribution

1,341.3
Mean
1,363.7
Median
121
Std dev
1,152.6
10th percentile
1,474.5
90th percentile
953.4
Min
953.41,527
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.10 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
251QwQ 32B· LMArena data1,335.5
252Nova 2 Lite· LMArena data1,334.8
253Llama 3.1 405B InstructFP8· LMArena data1,333.7
254Step-2 16K Exp202412· LMArena data1,333.5
255Olmo 3.1 32B Instruct· LMArena data1,329.2
256Yi Lightning· LMArena data1,328.3
257Llama 3.3 Nemotron Super 49B v1· LMArena data1,327.7
258Llama 4 Maverick· LMArena data1,327
259Qwen3 30B A3B· LMArena data1,326.5
260Hunyuan Large2025-02-10· LMArena data1,326.3
261Claude Haiku 3.520241022· LMArena data1,324.8
262GPT-4 Turbo2024-04-09· LMArena data1,324.6
263Gemini 1.5 Pro001· LMArena data1,324.2
264DeepSeek V2.51210· LMArena data1,323.4
265Ring Flash 2.0· LMArena data1,322.3
266Claude 3 Opus20240229· LMArena data1,322.3
267GPT-4.1 nano2025-04-14· LMArena data1,322.2
268Llama 4 Scout· LMArena data1,321.4
269Molmo 2 8B· LMArena data1,321
270Step-1o Turbo202506· LMArena data1,319.7
271GLM 4 Plus· LMArena data1,319.4
272Llama 3.3 70B Instruct· LMArena data1,318.1
273Qwen Max0919· LMArena data1,318.1
274Gemma 3n E4B IT· LMArena data1,317.6
275GPT-4o mini2024-07-18· LMArena data1,317.6
276GPT OSS 20b· LMArena data1,317.5
277Qwen2.5-Plus1127· LMArena data1,314.6
278Mistral Large 22407· LMArena data1,314.5
279Athene V2 Chat· LMArena data1,314.1
280Nemotron 3 Nano 30B A3BBF16· LMArena data1,313.5
281GPT-4 Turbo Preview0125· LMArena data1,313.3
282GPT-4 Turbo Preview1106· LMArena data1,313
283Hunyuan Standard2025-02-10· LMArena data1,311.3
284Gemini 1.5 Flash002· LMArena data1,309.4
285Grok 2 mini2024-08-13· LMArena data1,308.4
286DeepSeek V2.5· LMArena data1,307.2
287Athene 70B0725· LMArena data1,306.7
288Olmo 3 32B Think· LMArena data1,305.8
289Mistral Large2411· LMArena data1,305.6
290Mercury· LMArena data1,305.3
291Magistral Medium2506· LMArena data1,304.9
292Granite 4.1 8B· LMArena data1,304.9
293Gemma 3 4B IT· LMArena data1,303.3
294Qwen2.5 72B Instruct· LMArena data1,302.9
295Mistral Small 3.1 24B Instruct2503· LMArena data1,302.6
296Llama 3.1 Nemotron 70B Instruct· LMArena data1,298.8
297Hunyuan Large Vision· LMArena data1,293.8
298Llama 3.1 70B Instruct· LMArena data1,293.4
299Nova Pro 1.0· LMArena data1,290.1
300Gemma 2 27B IT· LMArena data1,289.6

Showing 251–300 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)