Skip to content

Benchmark

Arena ELO - Non English

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.20

Score distribution

1,288.7
Mean
1,334.4
Median
160.8
Std dev
1,032.5
10th percentile
1,449.6
90th percentile
781.3
Min
781.31,534.5
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.20 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
51Claude Opus 4.8· LMArena data1,443.8
52Doubao Seed 2.0 Pro· LMArena data1,441.8
53GPT-5.6 TerraxHigh· LMArena data1,441.8
54Grok 4.20 Multi-Agent0309, beta· LMArena data1,440.9
55Grok 4.20Reasoning, 0309, beta· LMArena data1,440.9
56Grok 4.5· LMArena data1,439.9
57Claude Sonnet 4.6· LMArena data1,439.4
58DeepSeek V4 Pro· LMArena data1,438.6
59Gemini 3 FlashMinimal· LMArena data1,437.9
60Claude Opus 4.520251101· LMArena data1,437.3
61Qwen3.6 Max Preview· LMArena data1,437
62GPT-6.1 SolMax· LMArena data1,436
63Inkling FP4· LMArena data1,435.1
64Claude Opus 4.5High, 32K, 20251101· LMArena data1,435
65ERNIE 5.00110· LMArena data1,435
66ERNIE-5.0-Preview1203· LMArena data1,434.9
67nova-experimental-chat26-02-10· LMArena data1,434.3
68Grok 4.20 Beta 1· LMArena data1,433
69Kimi K2.5Thinking· LMArena data1,432.1
70DeepSeek-V4-Pro PreviewHigh· LMArena data1,431.4
71GPT 5.1High· LMArena data1,431.1
72MiMo V2.6 Flash· LMArena data1,430.6
73Qwen3.5 397B A17B· LMArena data1,430.5
74GLM 5· LMArena data1,430.5
75Claude Sonnet 5High· LMArena data1,430.3
76GPT-6 AstraMax· LMArena data1,430.3
77Gemma 4 31B IT· LMArena data1,430.2
78Step 5 Preview· LMArena data1,429.5
79Qwen3 Max Preview· LMArena data1,429.3
80DeepSeek V4 ProHigh, 20260813· LMArena data1,429
81Nemotron 3 UltraNVFP4· LMArena data1,428.9
82Gemini 3.5 Flash Lite· LMArena data1,427.8
83Hy3 (Deep Infra)· LMArena data1,427.4
84ERNIE-5.0-Preview1022· LMArena data1,426.9
85GLM 4.6· LMArena data1,426.4
86Grok 4.1Thinking· LMArena data1,425.4
87Claude Sonnet 4.520250929· LMArena data1,425
88Qwen 3.8 27B· LMArena data1,424.9
89GPT-5.2 Chat20260210· LMArena data1,424.7
90Qwen3.6 Plus· LMArena data1,424.1
91Grok 4.1· LMArena data1,422.3
92Gemma 4 26B A4B· LMArena data1,421.6
93Grok 4.6High· LMArena data1,421.3
94DeepSeek V4 Flash· LMArena data1,419.5
95ChatGPT-4o Latest20250326· LMArena data1,418.9
96Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,417.6
97GLM 4.7· LMArena data1,417.4
98GPT-5.5 Instant· LMArena data1,417.3
99GLM 4.5· LMArena data1,417.1
100GPT-5.6 LunaxHigh· LMArena data1,417

Showing 51–100 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)