Skip to content

Benchmark

Arena ELO - French

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

Models measured
262
Scores recorded
303
Sources
1
IQ weight
1.87

Score distribution

1,360.8
Mean
1,408.4
Median
127.5
Std dev
1,158.9
10th percentile
1,487.2
90th percentile
945.6
Min
945.61,562.9
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.87 for chat models, measured over 262 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
251Nova Micro 1.0· LMArena data1,237.9
252Nova Lite 1.0· LMArena data1,237.5
253Gemini 1.5 Flash-8B001· LMArena data1,234
254Llama 3 70B Instruct· LMArena data1,232.2
255Claude 3 Sonnet20240229· LMArena data1,229.1
256Phi 4· LMArena data1,224.3
257GPT 40314· LMArena data1,219.4
258Mistral Large2402· LMArena data1,209.8
259Command R+· LMArena data1,208.6
260Mistral Small 24B Instruct2501· LMArena data1,203.4
261Nemotron-4 340B Instruct· LMArena data1,196.6
262Claude 3 Haiku20240307· LMArena data1,194.8
263Gemma 2 9B IT· LMArena data1,190.1
264DeepSeek Coder V2 Instruct· LMArena data1,184.8
265Mistral Medium· LMArena data1,179.8
266Llama 3.1 8B Instruct· LMArena data1,177.5
267Reka Flash 21B20240226· LMArena data1,174.4
268Qwen2 72B Instruct· LMArena data1,170.4
269GPT 40613· LMArena data1,170.1
270Mixtral 8x7B Instruct v0.1· LMArena data1,166.2
271Mixtral 8x22B Instruct v0.1· LMArena data1,166.1
272Reka Flash 21BOnline, 20240226· LMArena data1,162.8
273Qwen1.5-72B-Chat· LMArena data1,158.9
274Llama 3 8B Instruct· LMArena data1,158.8
275Gemma 2 2B IT· LMArena data1,157.2
276Yi 1.5 34B Chat· LMArena data1,156.4
277Qwen1.5-110B-Chat· LMArena data1,151.2
278Gemini ProDev API· LMArena data1,145.1
279Phi-3 Small 8K Instruct· LMArena data1,134.8
280Snowflake Arctic Instruct· LMArena data1,131.8
281GPT-3.5 Turbo0125· LMArena data1,117.8
282Starling LM 7B Beta· LMArena data1,116.2
283Qwen1.5-14B-Chat· LMArena data1,116
284Phi-3 Medium 4K Instruct· LMArena data1,110.5
285Qwen1.5-32B-Chat· LMArena data1,100.8
286DBRX Instruct Preview· LMArena data1,096.1
287OpenChat 3.50106· LMArena data1,092.2
288Llama 2 70B Chat· LMArena data1,090.1
289Vicuna 33B· LMArena data1,085.8
290Starling LM 7B Alpha· LMArena data1,085
291Yi 34B Chat· LMArena data1,081.4
292Phi-3 Mini 4K Instruct· LMArena data1,076.1
293GPT-3.5 Turbo1106· LMArena data1,066.4
294Gemma 1.1 7B IT· LMArena data1,065.5
295Llama 2 13B Chat· LMArena data1,044.2
296Phi-3 Mini 128K Instruct· LMArena data1,039
297Mistral 7B Instruct v0.2· LMArena data1,036.9
298StripedHyena Nous 7B· LMArena data1,032.1
299Zephyr 7B Beta· LMArena data1,030.9
300Gemma 7B IT· LMArena data1,025.1

Showing 251–300 of 303 results.

Data credits: LMArena, CC BY 4.0 (licence)