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-chat13690
1092.23
-4.99 +4.1
1
claude-sonnet-4-5-20250929
claude-sonnet-4-5-202509298882
1087.66
-4.78 +6.46
1
qwen3-235b-a22b-2507-v1
qwen3-235b-a22b-2507-v16578
1081.14
-5.18 +6.93
2
claude-opus-4-1-20250805
claude-opus-4-1-2025080514807
1081.42
-4 +3.47
5
claude-sonnet-4-20250514
claude-sonnet-4-2025051418082
1067.98
-3.73 +3.73
5
claude-opus-4-20250514
claude-opus-4-2025051414004
1062.40
-4.76 +3.4
5
claude-sonnet-4-5-20250929 (Thinking)
claude-sonnet-4-5-20250929 (Thinking)8612
1060.82
-4.27 +4.76
5
claude-haiku-4-5-20251001
claude-haiku-4-5-202510014899
1059.92
-6.08 +7.17
6
gpt-4.1-2025-04-14
gpt-4.1-2025-04-1415411
1058.32
-3.77 +4.08
7
claude-opus-4-1-20250805 (Thinking)
claude-opus-4-1-20250805 (Thinking)13637
1052.08
-5 +4.69
10
gemini-3-pro-preview
gemini-3-pro-preview3602
1046.24
-6.43 +5.7
10
gemini-2.5-pro
gemini-2.5-pro4465
1043.79
-6.97 +7.92
11
claude-opus-4-20250514 (Thinking)
claude-opus-4-20250514 (Thinking)14646
1039.16
-4.24 +3.46
12
claude-sonnet-4-20250514 (Thinking)
claude-sonnet-4-20250514 (Thinking)17250
1034.69
-4.01 +3.68
12
claude-haiku-4-5-20251001 (Thinking)
claude-haiku-4-5-20251001 (Thinking)4650
1030.14
-7.24 +7.28
15
gemini-2.5-flash
gemini-2.5-flash4539
1017.09
-7.73 +6.5
16
19495
1018.28
-3.73 +3.82
18
llama4-maverick-instruct-basic
llama4-maverick-instruct-basic16487
1000.00
-5.84 +4.15
19
19285
986.45
-2.99 +3.73
20
deepseek-r1-0528
deepseek-r1-05285783
965.95
-6.15 +6.95
21
14340
947.38
-4.71 +4.99
* 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.
