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
51DeepSeek V4 Pro· LMArena data1,470.7
52GPT-5.6 TerraxHigh· LMArena data1,470.5
53GLM 5.1· LMArena data1,469.8
54Claude Opus 4.520251101· LMArena data1,469.8
55DeepSeek V4.1 Flash 0910Max· LMArena data1,469.1
56MiMo V2 Pro· LMArena data1,468.4
57Grok 4.20 Multi-Agent0309, beta· LMArena data1,467.4
58Qwen 3.8 27B· LMArena data1,466.1
59Claude Sonnet 4.6· LMArena data1,465.2
60Qwen3.5 397B A17B· LMArena data1,464.5
61Inkling FP4· LMArena data1,464.4
62Grok 3 Preview02-24· LMArena data1,463.1
63Claude Sonnet 5High· LMArena data1,463
64ERNIE-5.0-Preview1203· LMArena data1,461.6
65GLM 4.6· LMArena data1,461.5
66Gemma 4 26B A4B· LMArena data1,461.2
67Beam 1.0 501B A23B· LMArena data1,459.9
68Mistral Medium 3.12508· LMArena data1,459.3
69ERNIE 5.00110· LMArena data1,458.9
70Claude Sonnet 4.520250929· LMArena data1,458.3
71Claude Opus 4.5High, 32K, 20251101· LMArena data1,457.3
72GLM 5· LMArena data1,457
73LongCat Flash Chat· LMArena data1,455.7
74Grok 4.20 Beta 1· LMArena data1,455.4
75Qwen3.6 Plus· LMArena data1,455.3
76Mistral Large 3· LMArena data1,455.1
77Grok 4.5· LMArena data1,454.7
78GPT 5.2High· LMArena data1,453.9
79GPT-5.6 LunaxHigh· LMArena data1,453.7
80Kimi K2.5Thinking· LMArena data1,453.1
81Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,451.2
82Qwen3.6 Max Preview· LMArena data1,450.9
83Qwen3 VL 235B· LMArena data1,450.4
84DeepSeek-V4-Pro PreviewHigh· LMArena data1,450.2
85GPT 5.1High· LMArena data1,450
86Grok 4.1Thinking· LMArena data1,449.7
87MiMo V2.5· LMArena data1,449.7
88GPT-6.1 SolMax· LMArena data1,449.6
89nova-experimental-chat11-10· LMArena data1,449.2
90MiMo V2 Omni· LMArena data1,449.1
91Qwen3 Max Preview· LMArena data1,448.7
92Mistral Medium 3.5· LMArena data1,448.5
93Gemini 3.5 Flash Lite· LMArena data1,448.1
94Grok 4.6High· LMArena data1,447.9
95DeepSeek V3.1· LMArena data1,447.6
96Hy3 (Deep Infra)· LMArena data1,446.2
97ChatGPT-4o Latest20250326· LMArena data1,446
98Gemini 3 FlashMinimal· LMArena data1,445.7
99Qwen3 235B A22B Instruct 25072507· LMArena data1,444.9
100GPT-6.1 TerraMax· LMArena data1,444.8

Showing 51–100 of 303 results.

Data credits: LMArena, CC BY 4.0 (licence)