Skip to content

Benchmark

Arena ELO - Russian

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

Models measured
332
Scores recorded
387
Sources
1
IQ weight
1.42

Score distribution

1,322.5
Mean
1,355.7
Median
134.6
Std dev
1,109.2
10th percentile
1,466.9
90th percentile
936.6
Min
936.61,560.5
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.42 for chat models, measured over 332 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
201Grok 3 MiniHigh· LMArena data1,350.8
202Gemma 3 27B IT· LMArena data1,349.4
203GPT OSS 120B· LMArena data1,343.2
204Nova 2 Lite· LMArena data1,342.5
205Qwen 3 235B A22B· LMArena data1,342
206GLM 4.6v· LMArena data1,339.6
207Trinity Large Thinking· LMArena data1,338.2
208Qwen3 Next 80B A3Bthinking· LMArena data1,337.7
209INTELLECT-3· LMArena data1,337.5
210Granite 4.2 30B· LMArena data1,336.8
211Nemotron 3 Super 120B A12B· LMArena data1,335.9
212Gemma 3 12B IT· LMArena data1,335.2
213o4 Mini2025-04-16· LMArena data1,333.6
214GLM 4.7 Flash· LMArena data1,332.6
215Step3· LMArena data1,332.1
216Llama 3.3 Nemotron Super 49B v1.5· LMArena data1,332
217MiniMax M1· LMArena data1,330.1
218MiniMax M2· LMArena data1,328.8
219Gemini 2.0 Flash-Lite Preview02-05· LMArena data1,327.5
220Step-1o Turbo202506· LMArena data1,326.9
221DeepSeek V3· LMArena data1,324.1
222Step-2 16K Exp202412· LMArena data1,323.9
223GPT-4.1 mini2025-04-14· LMArena data1,323.9
224Qwen Plus0125· LMArena data1,323.2
225Mistral Small 3.22506· LMArena data1,323.2
226Trinity Large Preview· LMArena data1,323.1
227Gemini 1.5 Pro002· LMArena data1,319.7
228O1 Preview· LMArena data1,316.1
229Command A03-2025· LMArena data1,313.6
230Hunyuan TurboS20250226· LMArena data1,313.1
231GLM 4 Plus0111· LMArena data1,312.6
232Claude Sonnet 3.7Thinking 32K, 20250219· LMArena data1,311.3
233Qwen3 32B· LMArena data1,310.7
234Claude Sonnet 3.720250219· LMArena data1,310.2
235Ling Flash 2.0· LMArena data1,307.5
236Claude 3.5 Sonnet20241022· LMArena data1,305.3
237Hunyuan Turbo0110· LMArena data1,305.1
238Mercury 2· LMArena data1,304.4
239o3 MiniHigh· LMArena data1,304
240Nemotron 3 Nano 30B A3BBF16· LMArena data1,303.7
241o3 Mini· LMArena data1,300.9
242GPT-5 nanoHigh· LMArena data1,297.9
243GLM 4.5V· LMArena data1,297.5
244QwQ 32B· LMArena data1,297.2
245Gemma 3 4B IT· LMArena data1,293.8
246Qwen3 30B A3B· LMArena data1,290.8
247DeepSeek V2.51210· LMArena data1,289.3
248Gemma 3n E4B IT· LMArena data1,287.8
249Gemini 1.5 Flash002· LMArena data1,287.5
250Llama 4 Maverick· LMArena data1,286.4

Showing 201–250 of 387 results.

Data credits: LMArena, CC BY 4.0 (licence)