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
251Grok 22024-08-13· LMArena data1,285.7
252GPT 4o2024-05-13· LMArena data1,285.5
253Llama 3.1 Nemotron Ultra 253B v1· LMArena data1,283.7
254O1 Mini· LMArena data1,282.5
255Gemini Advanced0514· LMArena data1,282.4
256Claude 3.5 Sonnet20240620· LMArena data1,282.4
257Athene V2 Chat· LMArena data1,281.9
258Claude 3 Opus20240229· LMArena data1,279.9
259Granite 4.2 8B· LMArena data1,279.1
260GPT OSS 20b· LMArena data1,278
261Gemini 1.5 Pro001· LMArena data1,276.2
262GPT-4o mini2024-07-18· LMArena data1,275.2
263GLM 4 Plus· LMArena data1,275.1
264Qwen Max0919· LMArena data1,274
265GPT 4o2024-08-06· LMArena data1,271.8
266Qwen2.5-Plus1127· LMArena data1,270.6
267Llama 3.3 Nemotron Super 49B v1· LMArena data1,269.3
268Olmo 3.1 32B Instruct· LMArena data1,268.2
269Hunyuan Large2025-02-10· LMArena data1,265.9
270Llama 3.1 405B InstructFP8· LMArena data1,265.2
271Qwen2.5 72B Instruct· LMArena data1,263.9
272Mistral Small 3.1 24B Instruct2503· LMArena data1,263.2
273Llama 4 Scout· LMArena data1,262
274Grok 2 mini2024-08-13· LMArena data1,261.1
275GPT-4.1 nano2025-04-14· LMArena data1,261
276GPT-4 Turbo2024-04-09· LMArena data1,259.4
277Hunyuan Standard2025-02-10· LMArena data1,258.3
278Mistral Large 22407· LMArena data1,256.8
279Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data1,256.2
280Llama 3.1 405B InstructBF16· LMArena data1,256.1
281Yi Lightning· LMArena data1,255.1
282Olmo 3 32B Think· LMArena data1,253
283DeepSeek V2.5· LMArena data1,252.7
284Claude Haiku 3.520241022· LMArena data1,252.7
285Llama 3.3 70B Instruct· LMArena data1,251.8
286Mistral Large2411· LMArena data1,251.5
287GPT-4 Turbo Preview1106· LMArena data1,251.3
288Athene 70B0725· LMArena data1,250.4
289Reka Core20240904· LMArena data1,249.2
290Llama 3.1 Tulu 3 70B· LMArena data1,245.6
291Ring Flash 2.0· LMArena data1,245.5
292Hunyuan Large Vision· LMArena data1,242.8
293GPT-4 Turbo Preview0125· LMArena data1,240.9
294Granite 4.2 3B· LMArena data1,240.2
295Nova Pro 1.0· LMArena data1,239.8
296Gemini 1.5 Flash001· LMArena data1,239.3
297Granite 4.1 8B· LMArena data1,239.2
298Gemini 1.5 Flash-8B001· LMArena data1,235.7
299Gemma 2 27B IT· LMArena data1,233.8
300Llama 3.1 70B Instruct· LMArena data1,233.6

Showing 251–300 of 387 results.

Data credits: LMArena, CC BY 4.0 (licence)