Skip to content

Benchmark

Arena ELO Text Factuality - French

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

Models measured
82
Scores recorded
95
Sources
1
IQ weight
2.21

Score distribution

1,476.5
Mean
1,478.8
Median
28.3
Std dev
1,443.6
10th percentile
1,513.7
90th percentile
1,398.5
Min
1,398.51,546.9
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.21 for chat models, measured over 82 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51Claude Sonnet 4.520250929· LMArena data1,474.8
52MiMo V2.6 Flash· LMArena data1,474.5
53Grok 4.5· LMArena data1,474.5
54Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,474.4
55Qwen3.5 397B A17B· LMArena data1,473.9
56MiMo V2.5· LMArena data1,473.5
57Qwen 3.8 27B· LMArena data1,471.1
58Gemini 3 FlashMinimal· LMArena data1,470.8
59Grok 4.6High· LMArena data1,470.1
60Doubao Seed 2.0 Pro· LMArena data1,469.7
61GPT-6 LunaMax· LMArena data1,466
62Mistral Large 3· LMArena data1,463.1
63Gemini 3.5 Flash Lite· LMArena data1,462.6
64DeepSeek V4 Flash· LMArena data1,462.3
65Qwen3.6 Plus· LMArena data1,462.2
66Mistral Medium 3.12508· LMArena data1,461.4
67GPT-5.5 Instant· LMArena data1,461
68Kimi K2.5Thinking· LMArena data1,457.9
69DeepSeek-V4-Flash PreviewHigh· LMArena data1,457.5
70Gemini 2.5 Flash· LMArena data1,456.3
71ERNIE 5.00110· LMArena data1,456
72DeepSeek V3.2· LMArena data1,455.5
73GPT 5.1· LMArena data1,454.7
74DeepSeek-V4-Pro PreviewHigh· LMArena data1,454.4
75Claude Opus 4.120250805· LMArena data1,453.4
76Grok 4.1Thinking· LMArena data1,450.5
77Gemini 3.1 Flash Lite· LMArena data1,450.1
78MiniMax M3· LMArena data1,449.7
79MiniMax M2.7· LMArena data1,449.4
80Step 3.5 Flash· LMArena data1,449
81Inkling Small· LMArena data1,448.8
82Qwen3 235B A22B Instruct 25072507· LMArena data1,448.4
83DeepSeek V3.2Thinking· LMArena data1,448.2
84Grok 4.3· LMArena data1,445.9
85Grok 4.1· LMArena data1,444.5
86GPT 5.4 NanoHigh· LMArena data1,443.6
87Kimi K2 Thinking· LMArena data1,441.2
88Qwen3.5 Flash· LMArena data1,436.1
89Grok 4.1 FastReasoning· LMArena data1,434.2
90Claude Haiku 4.520251001· LMArena data1,430.8
91MiMo V2 FlashNon-Thinking· LMArena data1,429
92Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,419.2
93Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data1,414.9
94Trinity Large Preview· LMArena data1,412.5
95MiniMax M2.5· LMArena data1,398.5

Showing 51–95 of 95 results.

Data credits: LMArena, CC BY 4.0 (licence)