Skip to content

Benchmark

Arena ELO - Industry Software and It Services

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.29

Score distribution

1,322.6
Mean
1,374.8
Median
161.3
Std dev
1,081.5
10th percentile
1,486.1
90th percentile
719.3
Min
719.31,552.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.29 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
301Claude 3 Opus20240229· LMArena data1,258.8
302Granite 4.0 H Small· LMArena data1,252.9
303GPT-4 Turbo Preview1106· LMArena data1,250.1
304GPT-4 Turbo Preview0125· LMArena data1,249.5
305Mistral Small 24B Instruct2501· LMArena data1,248.1
306Llama 3.1 Tulu 3 70B· LMArena data1,241.7
307Gemini 1.5 Flash001· LMArena data1,241.5
308Nova Lite 1.0· LMArena data1,241.1
309DeepSeek Coder V2 Instruct· LMArena data1,238.7
310Jamba 1.5 Large· LMArena data1,236.4
311Reka Core20240904· LMArena data1,231.7
312Llama 3.1 Nemotron 51B Instruct· LMArena data1,231.1
313Gemma 2 27B IT· LMArena data1,227.9
314Phi 4· LMArena data1,227.8
315Gemini 1.5 Flash-8B001· LMArena data1,227.2
316GLM 40520· LMArena data1,225.7
317Hunyuan Standard 256K· LMArena data1,222.3
318Claude 3 Sonnet20240229· LMArena data1,220.8
319Nemotron-4 340B Instruct· LMArena data1,215.3
320Nova Micro 1.0· LMArena data1,215
321OLMo 2 32B Instruct0325· LMArena data1,213.8
322Ministral 8B2410· LMArena data1,213.2
323Aya Expanse 32B· LMArena data1,212.4
324Gemma 2 9B IT SimPO· LMArena data1,211.4
325Llama 3 70B Instruct· LMArena data1,204.3
326Reka Flash20240904· LMArena data1,198.7
327Command R+08-2024· LMArena data1,197
328Gemma 2 9B IT· LMArena data1,196.5
329Llama 3.1 8B Instruct· LMArena data1,194.9
330GPT 40314· LMArena data1,194.7
331Claude 3 Haiku20240307· LMArena data1,193.9
332Qwen2 72B Instruct· LMArena data1,192.9
333Llama 3.1 Tulu 3 8B· LMArena data1,191.5
334Jamba 1.5 Mini· LMArena data1,188.8
335Command R+· LMArena data1,185.8
336Aya Expanse 8B· LMArena data1,176.8
337Qwen1.5-110B-Chat· LMArena data1,174.3
338Yi 1.5 34B Chat· LMArena data1,171.9
339Mistral Large2402· LMArena data1,170.6
340GPT 40613· LMArena data1,168.7
341Reka Flash 21BOnline, 20240226· LMArena data1,164.7
342InternLM2.5 20B Chat· LMArena data1,163.6
343Granite 3.1 8B Instruct· LMArena data1,161.9
344Qwen1.5-72B-Chat· LMArena data1,160.8
345Llama 3 8B Instruct· LMArena data1,156
346Reka Flash 21B20240226· LMArena data1,155.4
347QwQ 32B Preview· LMArena data1,154.2
348Mixtral 8x22B Instruct v0.1· LMArena data1,154.1
349Mistral Medium· LMArena data1,153.7
350Qwen1.5-32B-Chat· LMArena data1,146.5

Showing 301–350 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)