Skip to content

Benchmark

Arena ELO - Hard Prompts

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.24

Score distribution

1,308.5
Mean
1,356.7
Median
162.3
Std dev
1,065.4
10th percentile
1,480.3
90th percentile
727.9
Min
727.91,550.7
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.24 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
51Kimi K3.1 ProMax· LMArena data1,469
52GPT 5.4· LMArena data1,468.5
53MiMo V2.6 Flash· LMArena data1,467.9
54GPT-6.1 SolMax· LMArena data1,467.9
55Gemini 3 Flash· LMArena data1,464.8
56Step 5 Preview· LMArena data1,463.5
57GPT-6 AstraMax· LMArena data1,463
58Claude Sonnet 4.520250929· LMArena data1,462
59Grok 4.5· LMArena data1,462
60DeepSeek V4 Pro· LMArena data1,461.5
61Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,461.3
62Claude Sonnet 5High· LMArena data1,461.1
63nova-experimental-chat26-02-10· LMArena data1,459
64Qwen3.7 Plus· LMArena data1,458.8
65Qwen 3.8 27B· LMArena data1,457.8
66MiMo V2 Pro· LMArena data1,457.7
67GPT 5.1High· LMArena data1,456.7
68Qwen3.6 Max Preview· LMArena data1,456.2
69DeepSeek V4 ProHigh, 20260813· LMArena data1,456.1
70Gemini 2.5 Pro· LMArena data1,454.9
71Doubao Seed 2.0 Pro· LMArena data1,453.5
72Kimi K2.5Thinking· LMArena data1,453.2
73Inkling FP4· LMArena data1,452.9
74Nemotron 3 UltraNVFP4· LMArena data1,452.2
75DeepSeek-V4-Pro PreviewHigh· LMArena data1,452
76GLM 5· LMArena data1,451.8
77GPT-5.6 LunaxHigh· LMArena data1,451.6
78Grok 4.20Reasoning, 0309, beta· LMArena data1,451.5
79GPT-6.1 TerraMax· LMArena data1,450.5
80MiMo V2.5· LMArena data1,450.2
81Grok 4.20 Multi-Agent0309, beta· LMArena data1,448.6
82Qwen3 Max Preview· LMArena data1,448.5
83Qwen3.6 Plus· LMArena data1,448.3
84Grok 4.6High· LMArena data1,448.2
85Hy3 (Deep Infra)· LMArena data1,448.2
86Qwen3.5 397B A17B· LMArena data1,447.6
87Gemma 4 31B IT· LMArena data1,447.2
88ERNIE 5.00110· LMArena data1,445.1
89GPT-5.2 Chat20260210· LMArena data1,445
90MiMo V2 Omni· LMArena data1,444.3
91Kimi K2.5Instant· LMArena data1,443.6
92DeepSeek V4 Flash· LMArena data1,443.3
93GLM 4.7· LMArena data1,442.9
94Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,442.3
95Beam 1.0 501B A23B· LMArena data1,442.2
96Claude Opus 4.120250805· LMArena data1,442.1
97MiniMax M3· LMArena data1,441.1
98Gemini 3.5 Flash Lite· LMArena data1,440.9
99Grok 4.20 Beta 1· LMArena data1,440.8
100GLM 4.6· LMArena data1,440.3

Showing 51–100 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)