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
301Qwen 2.5 Coder 32B Instruct· LMArena data1,228.1
302Command R+08-2024· LMArena data1,227.3
303Aya Expanse 32B· LMArena data1,227
304Llama 3.1 Nemotron 70B Instruct· LMArena data1,226.8
305Claude 3 Sonnet20240229· LMArena data1,226.6
306Gemma 2 9B IT SimPO· LMArena data1,224.5
307Magistral Medium2506· LMArena data1,222.3
308Nemotron-4 340B Instruct· LMArena data1,219.3
309Mistral Small 24B Instruct2501· LMArena data1,216
310Nova Lite 1.0· LMArena data1,215.7
311GLM 40520· LMArena data1,213.2
312Reka Flash20240904· LMArena data1,211.5
313Phi 4· LMArena data1,208.6
314Command R+· LMArena data1,205.9
315Claude 3 Haiku20240307· LMArena data1,204.1
316Granite 4.0 H Small· LMArena data1,200.9
317Gemma 2 9B IT· LMArena data1,200.4
318Aya Expanse 8B· LMArena data1,197.2
319Jamba 1.5 Large· LMArena data1,195.7
320GPT 40314· LMArena data1,195.3
321Ministral 8B2410· LMArena data1,195.1
322Olmo 3.1 32B Think· LMArena data1,193.2
323Llama 3.1 Tulu 3 8B· LMArena data1,193.1
324DeepSeek Coder V2 Instruct· LMArena data1,187.7
325OLMo 2 32B Instruct0325· LMArena data1,186.8
326Llama 3.1 Nemotron 51B Instruct· LMArena data1,186.4
327Gemini ProDev API· LMArena data1,186
328Nova Micro 1.0· LMArena data1,185.2
329Mistral Large2402· LMArena data1,179.5
330GPT 40613· LMArena data1,173.1
331Qwen2 72B Instruct· LMArena data1,168.7
332Mistral Medium· LMArena data1,167.6
333Reka Flash 21BOnline, 20240226· LMArena data1,163.3
334Reka Flash 21B20240226· LMArena data1,159.8
335Llama 3 70B Instruct· LMArena data1,158.7
336Hunyuan Standard 256K· LMArena data1,158.6
337Llama 3.1 8B Instruct· LMArena data1,158.3
338Mixtral 8x22B Instruct v0.1· LMArena data1,157.7
339Jamba 1.5 Mini· LMArena data1,156.2
340WizardLM 70B· LMArena data1,155.3
341Phi-3 Medium 4K Instruct· LMArena data1,144.9
342GPT-3.5 Turbo0125· LMArena data1,123.2
343Qwen1.5-110B-Chat· LMArena data1,118.3
344Gemma 2 2B IT· LMArena data1,118.1
345Zephyr ORPO 141B A35B v0.1· LMArena data1,115.8
346QwQ 32B Preview· LMArena data1,114.5
347Phi-3 Small 8K Instruct· LMArena data1,110.5
348Llama 3 8B Instruct· LMArena data1,109.2
349InternLM2.5 20B Chat· LMArena data1,106.9
350OpenChat 3.5· LMArena data1,106.6

Showing 301–350 of 387 results.

Data credits: LMArena, CC BY 4.0 (licence)