Skip to content

Benchmark

Arena ELO - Hard Prompts

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.24

Score distribution

1,308.5
Mean
1,356.7
Median
162.3
Std dev
1,065.4
10th percentile
1,480.3
90th percentile
727.9
Min
727.91,550.7
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.24 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
351Qwen1.5-32B-Chat· LMArena data1,130.4
352Phi-3 Medium 4K Instruct· LMArena data1,126.5
353Starling LM 7B Beta· LMArena data1,118.4
354Zephyr ORPO 141B A35B v0.1· LMArena data1,115.6
355Mixtral 8x7B Instruct v0.1· LMArena data1,114.6
356Qwen1.5-14B-Chat· LMArena data1,113.1
357DBRX Instruct Preview· LMArena data1,112.9
358Gemma 2 2B IT· LMArena data1,112.4
359Gemini ProDev API· LMArena data1,109.2
360GPT-3.5 Turbo0125· LMArena data1,107.8
361Tulu 2 DPO 70B· LMArena data1,104.3
362Yi 34B Chat· LMArena data1,103.5
363Phi-3 Small 8K Instruct· LMArena data1,100.2
364GPT-3.5 Turbo1106· LMArena data1,098.5
365Llama 3.2 3B· LMArena data1,094.5
366Gemini Pro· LMArena data1,093.7
367Granite 3.0 8B Instruct· LMArena data1,091.6
368Phi-3 Mini 4K InstructJune 2024· LMArena data1,087.4
369OpenChat 3.50106· LMArena data1,084.9
370WizardLM 70B· LMArena data1,079.1
371Starling LM 7B Alpha· LMArena data1,075.3
372Granite 3.0 2B Instruct· LMArena data1,073.2
373Llama 2 70B Chat· LMArena data1,072.9
374Phi-3 Mini 4K Instruct· LMArena data1,072.1
375OpenHermes 2.5 Mistral 7B· LMArena data1,071.5
376Snowflake Arctic Instruct· LMArena data1,071.4
377Gemma 1.1 7B IT· LMArena data1,070.6
378DeepSeek LLM 67B Chat· LMArena data1,069.7
379Mistral 7B Instruct v0.2· LMArena data1,067.1
380Vicuna 33B· LMArena data1,066.8
381OpenChat 3.5· LMArena data1,065.4
382Qwen1.5-7B-Chat· LMArena data1,065.3
383SOLAR 10.7B Instruct v1.0· LMArena data1,062
384Dolphin 2.2.1 Mistral 7B· LMArena data1,054.5
385Nous Hermes 2 Mixtral 8x7B DPO· LMArena data1,053.3
386SmolLM2 1.7B Instruct· LMArena data1,053
387Code Llama 70B Instruct· LMArena data1,052.1
388MPT 30B Chat· LMArena data1,051.5
389Llama 2 13B Chat· LMArena data1,050.8
390Llama 2 70B SteerLM Chat· LMArena data1,046.6
391Llama 3.2 1B Instruct· LMArena data1,044.2
392Gemma 7B IT· LMArena data1,041.5
393Code Llama 34B Instruct· LMArena data1,032
394Phi-3 Mini 128K Instruct· LMArena data1,027.9
395Qwen-14B-Chat· LMArena data1,027
396Zephyr 7B Beta· LMArena data1,026.5
397Zephyr 7B Alpha· LMArena data1,023.7
398Vicuna 13B· LMArena data1,023.5
399WizardLM 13B· LMArena data1,017.6
400Llama 2 7B Chat· LMArena data1,009.1

Showing 351–400 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)