Showdown Leaderboard - LLMs
Real people. Real conversations. Real rankings.
Showdown ranks AI models based on how they perform in real-world use— not synthetic tests or lab settings. Votes are blind, optional, and organic, so rankings reflect authentic preferences.Methodology & Technical Report→0 promptsReal conversation prompts compared across models through pairwise votes.
0 usersFrom 80+ countries and 70+ languages, spanning all backgrounds and professions.
SEAL Leaderboard - LLMs
RANK ↑
MODEL ↑↓
VOTES ↑↓
SCORE ↑↓
1
gpt-5-chat
gpt-5-chat13799
1092.66
-4.43 +3.98
1
claude-sonnet-4-5-20250929
claude-sonnet-4-5-2025092910320
1086.46
-4.86 +4.36
2
qwen3-235b-a22b-2507-v1
qwen3-235b-a22b-2507-v19304
1080.76
-3.6 +5
2
claude-opus-4-1-20250805
claude-opus-4-1-2025080514828
1080.26
-4.71 +3.3
5
claude-sonnet-4-20250514
claude-sonnet-4-2025051419958
1067.73
-4.03 +3.63
5
claude-sonnet-4-5-20250929 (Thinking)
claude-sonnet-4-5-20250929 (Thinking)10061
1063.58
-4.76 +4.06
5
claude-opus-4-20250514
claude-opus-4-2025051414025
1061.72
-3.89 +3.8
5
claude-haiku-4-5-20251001
claude-haiku-4-5-202510014935
1060.19
-7.11 +6.31
6
gpt-4.1-2025-04-14
gpt-4.1-2025-04-1415423
1058.20
-3.61 +3.25
8
claude-opus-4-1-20250805 (Thinking)
claude-opus-4-1-20250805 (Thinking)13654
1050.95
-4.79 +4.01
10
gemini-3-pro-preview
gemini-3-pro-preview5208
1046.80
-6.31 +5.68
11
gemini-2.5-pro
gemini-2.5-pro6951
1038.84
-5.35 +7.05
11
claude-opus-4-20250514 (Thinking)
claude-opus-4-20250514 (Thinking)14705
1038.74
-3.49 +4
12
claude-sonnet-4-20250514 (Thinking)
claude-sonnet-4-20250514 (Thinking)17333
1033.93
-2.81 +3.34
12
claude-haiku-4-5-20251001 (Thinking)
claude-haiku-4-5-20251001 (Thinking)4692
1029.15
-5.72 +6.51
16
21548
1017.58
-2.97 +2.85
16
gemini-2.5-flash
gemini-2.5-flash7343
1013.12
-7.02 +6.44
18
llama4-maverick-instruct-basic
llama4-maverick-instruct-basic16505
1000.00
-4.15 +3.53
19
21158
986.81
-3.42 +3.65
20
deepseek-r1-0528
deepseek-r1-05287354
963.14
-6.29 +5.66
21
16354
946.69
-3.68 +3.72
* This model’s API does not consistently return Markdown-formatted responses. Since raw outputs are used in head-to-head comparisons, this may affect its ranking.
