Skip to content

Benchmark

Arena ELO Text Factuality - Industry Medicine and Healthcare

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

Models measured
113
Scores recorded
128
Sources
1
IQ weight
2.07

Score distribution

1,475.5
Mean
1,474.9
Median
28.4
Std dev
1,444.2
10th percentile
1,511.8
90th percentile
1,365.9
Min
1,365.91,540.5
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.07 for chat models, measured over 113 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51Inkling FP4· LMArena data1,482.5
52Qwen3.5 397B A17B· LMArena data1,481.9
53MiMo V2.6 Flash· LMArena data1,481.5
54GPT-5.6 TerraxHigh· LMArena data1,481.5
55Claude Sonnet 5.5xHigh· LMArena data1,481.2
56ERNIE 5.00110· LMArena data1,481
57Grok 4.5· LMArena data1,479.9
58Claude Sonnet 5High· LMArena data1,479.3
59GPT-5.6 SolxHigh· LMArena data1,478.5
60Claude Sonnet 4.6· LMArena data1,478.4
61Hy3 (Deep Infra)· LMArena data1,477.6
62GLM 5· LMArena data1,477.5
63GPT-6 AstraMax· LMArena data1,477.1
64Gemini 3 FlashMinimal· LMArena data1,474.9
65DeepSeek V4 Flash· LMArena data1,474.9
66Nemotron 3 UltraNVFP4· LMArena data1,474.7
67Grok 4.20 Beta 1· LMArena data1,474
68Mistral Large 3· LMArena data1,473.6
69GPT 5.2· LMArena data1,472.9
70Mistral Medium 3.12508· LMArena data1,472.2
71Step 5 Preview· LMArena data1,472.1
72Qwen3 235B A22B Instruct 25072507· LMArena data1,471.9
73Claude Opus 4.520251101· LMArena data1,471.9
74Claude Sonnet 4.520250929· LMArena data1,470.1
75Grok 4.6High· LMArena data1,469.3
76DeepSeek V3.2· LMArena data1,468.7
77nova-experimental-chat11-10· LMArena data1,468.3
78Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,467.6
79Claude Opus 4.5High, 32K, 20251101· LMArena data1,467.5
80GPT 5.2High· LMArena data1,466.9
81Qwen3.6 Plus· LMArena data1,466.6
82Kimi K2.5Thinking· LMArena data1,466.3
83DeepSeek-V4-Flash PreviewHigh· LMArena data1,466.2
84Qwen 3.8 27B· LMArena data1,463.9
85Qwen3.5 27B· LMArena data1,463.8
86GPT-5.2 Chat20260210· LMArena data1,463.6
87MiMo V2 Pro· LMArena data1,463.3
88Gemini 3.5 Flash Lite· LMArena data1,462.4
89Gemini 3.1 Flash Lite· LMArena data1,461
90MiniMax M3· LMArena data1,460.2
91Grok 4.1Thinking· LMArena data1,460
92DeepSeek V3.2Thinking· LMArena data1,459.6
93Grok 4.1· LMArena data1,459.1
94Qwen3.5 35B A3B· LMArena data1,458.8
95GPT-5.5 Instant· LMArena data1,458.2
96Claude Opus 4.120250805· LMArena data1,458.2
97MiMo V2 Omni· LMArena data1,458.1
98Qwen3.5 122B-A10B· LMArena data1,457.8
99Gemini 2.5 Flash· LMArena data1,457.2
100Doubao Seed 2.0 Pro· LMArena data1,456.6

Showing 51–100 of 128 results.

Data credits: LMArena, CC BY 4.0 (licence)