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
301GPT-4 Turbo Preview1106· LMArena data1,241.1
302Llama 3.1 70B Instruct· LMArena data1,241.1
303Granite 4.0 H Small· LMArena data1,241.1
304GPT-4 Turbo Preview0125· LMArena data1,238.5
305Mistral Small 24B Instruct2501· LMArena data1,233
306Llama 3.1 Tulu 3 70B· LMArena data1,219.9
307Nova Lite 1.0· LMArena data1,219.7
308Gemini 1.5 Flash001· LMArena data1,219.5
309Phi 4· LMArena data1,219.5
310Hunyuan Standard 256K· LMArena data1,218
311Jamba 1.5 Large· LMArena data1,217.6
312GLM 40520· LMArena data1,211.9
313Reka Core20240904· LMArena data1,210.9
314Gemini 1.5 Flash-8B001· LMArena data1,209.3
315OLMo 2 32B Instruct0325· LMArena data1,207.6
316DeepSeek Coder V2 Instruct· LMArena data1,206.7
317Llama 3.1 Nemotron 51B Instruct· LMArena data1,203.4
318Nemotron-4 340B Instruct· LMArena data1,201.4
319GPT 40314· LMArena data1,200
320Gemma 2 27B IT· LMArena data1,198.3
321Claude 3 Sonnet20240229· LMArena data1,196.9
322Gemma 2 9B IT SimPO· LMArena data1,196
323Llama 3 70B Instruct· LMArena data1,194.8
324Aya Expanse 32B· LMArena data1,193
325Ministral 8B2410· LMArena data1,191.1
326Nova Micro 1.0· LMArena data1,191
327Qwen2 72B Instruct· LMArena data1,190.5
328Command R+08-2024· LMArena data1,185.5
329Reka Flash20240904· LMArena data1,181.3
330Llama 3.1 8B Instruct· LMArena data1,174.6
331GPT 40613· LMArena data1,174.3
332Llama 3.1 Tulu 3 8B· LMArena data1,174.2
333Claude 3 Haiku20240307· LMArena data1,173.9
334Gemma 2 9B IT· LMArena data1,171.3
335Qwen1.5-110B-Chat· LMArena data1,167.8
336Mistral Large2402· LMArena data1,167
337Jamba 1.5 Mini· LMArena data1,166.7
338Command R+· LMArena data1,166.4
339QwQ 32B Preview· LMArena data1,162.5
340Yi 1.5 34B Chat· LMArena data1,159.7
341InternLM2.5 20B Chat· LMArena data1,157.9
342Aya Expanse 8B· LMArena data1,155.8
343Mixtral 8x22B Instruct v0.1· LMArena data1,149.6
344Mistral Medium· LMArena data1,149.2
345Qwen1.5-72B-Chat· LMArena data1,147.8
346Granite 3.1 8B Instruct· LMArena data1,145.3
347Granite 3.1 2B Instruct· LMArena data1,138
348Reka Flash 21BOnline, 20240226· LMArena data1,136.8
349Reka Flash 21B20240226· LMArena data1,136.4
350Llama 3 8B Instruct· LMArena data1,133.2

Showing 301–350 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)