Skip to content

Benchmark

Arena ELO - Industry Business and Management and Financial Operations

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

Models measured
359
Scores recorded
416
Sources
1
IQ weight
1.28

Score distribution

1,295.6
Mean
1,344.1
Median
158.7
Std dev
1,054.4
10th percentile
1,457
90th percentile
734.2
Min
734.21,520.2
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.28 for chat models, measured over 359 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
301GPT-4 Turbo2024-04-09· LMArena data1,222.4
302Gemma 2 9B IT SimPO· LMArena data1,222.3
303Claude 3 Opus20240229· LMArena data1,222
304Nova Lite 1.0· LMArena data1,218.8
305Reka Core20240904· LMArena data1,217.1
306GPT-4 Turbo Preview0125· LMArena data1,216.6
307Llama 3.1 Nemotron 51B Instruct· LMArena data1,215.4
308Command R+08-2024· LMArena data1,215.1
309Granite 4.0 H Small· LMArena data1,212.4
310GPT-4 Turbo Preview1106· LMArena data1,211
311Gemini 1.5 Flash-8B001· LMArena data1,210.8
312Aya Expanse 32B· LMArena data1,210
313Gemma 2 27B IT· LMArena data1,207.6
314Mistral Small 24B Instruct2501· LMArena data1,205.3
315Phi 4· LMArena data1,198.7
316Llama 3.1 Tulu 3 8B· LMArena data1,196.4
317Jamba 1.5 Large· LMArena data1,196.2
318Reka Flash20240904· LMArena data1,191.3
319Command R+· LMArena data1,190.4
320Gemma 2 9B IT· LMArena data1,189.9
321Nemotron-4 340B Instruct· LMArena data1,184.8
322Nova Micro 1.0· LMArena data1,181.9
323OLMo 2 32B Instruct0325· LMArena data1,181
324Hunyuan Standard 256K· LMArena data1,177.7
325GLM 40520· LMArena data1,176.7
326Aya Expanse 8B· LMArena data1,176.2
327Claude 3 Sonnet20240229· LMArena data1,174
328Ministral 8B2410· LMArena data1,172.7
329Llama 3 70B Instruct· LMArena data1,172.3
330Llama 3.1 8B Instruct· LMArena data1,165.9
331Claude 3 Haiku20240307· LMArena data1,157.5
332Jamba 1.5 Mini· LMArena data1,157.4
333DeepSeek Coder V2 Instruct· LMArena data1,155.1
334Qwen2 72B Instruct· LMArena data1,146.3
335GPT 40314· LMArena data1,145.1
336Granite 3.1 8B Instruct· LMArena data1,142
337Qwen1.5-110B-Chat· LMArena data1,140.2
338Granite 3.1 2B Instruct· LMArena data1,139.2
339Reka Flash 21BOnline, 20240226· LMArena data1,135.4
340Yi 1.5 34B Chat· LMArena data1,134.7
341Gemini ProDev API· LMArena data1,130.7
342Mistral Large2402· LMArena data1,130.2
343Llama 3 8B Instruct· LMArena data1,127.1
344Reka Flash 21B20240226· LMArena data1,126.9
345Gemma 2 2B IT· LMArena data1,125.6
346GPT 40613· LMArena data1,121.7
347Mistral Medium· LMArena data1,119.5
348Mixtral 8x22B Instruct v0.1· LMArena data1,118.6
349Qwen1.5-72B-Chat· LMArena data1,116.9
350InternLM2.5 20B Chat· LMArena data1,115.1

Showing 301–350 of 416 results.

Data credits: LMArena, CC BY 4.0 (licence)