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
1092.97
-3.98 +4.03
1
claude-sonnet-4-5-20250929
claude-sonnet-4-5-202509299090
1088.98
-5.82 +4.26
2
claude-opus-4-1-20250805
claude-opus-4-1-2025080514828
1081.60
-4.41 +3.82
2
qwen3-235b-a22b-2507-v1
qwen3-235b-a22b-2507-v17039
1081.00
-4.57 +5.61
5
claude-sonnet-4-20250514
claude-sonnet-4-2025051418370
1069.00
-4.24 +3.14
5
claude-opus-4-20250514
claude-opus-4-2025051414025
1063.13
-4.19 +4.29
5
claude-sonnet-4-5-20250929 (Thinking)
claude-sonnet-4-5-20250929 (Thinking)8865
1062.69
-5.06 +3.87
5
claude-haiku-4-5-20251001
claude-haiku-4-5-202510014935
1060.94
-8.86 +6.43
6
gpt-4.1-2025-04-14
gpt-4.1-2025-04-1415423
1059.22
-4.12 +3.94
8
claude-opus-4-1-20250805 (Thinking)
claude-opus-4-1-20250805 (Thinking)13654
1052.62
-4.15 +4.34
8
gemini-3-pro-preview
gemini-3-pro-preview3857
1048.03
-7.3 +7.68
11
gemini-2.5-pro
gemini-2.5-pro4908
1041.26
-5.59 +6.6
11
claude-opus-4-20250514 (Thinking)
claude-opus-4-20250514 (Thinking)14705
1039.83
-3.6 +3.82
12
claude-sonnet-4-20250514 (Thinking)
claude-sonnet-4-20250514 (Thinking)17333
1035.16
-3.73 +3.77
12
claude-haiku-4-5-20251001 (Thinking)
claude-haiku-4-5-20251001 (Thinking)4692
1030.16
-6.38 +7.64
15
gemini-2.5-flash
gemini-2.5-flash5076
1017.37
-6.97 +6.64
16
19811
1018.78
-3.67 +3.27
18
llama4-maverick-instruct-basic
llama4-maverick-instruct-basic16505
1000.00
-3.14 +3.49
19
19605
987.13
-4.01 +2.99
20
deepseek-r1-0528
deepseek-r1-05286001
964.13
-4.81 +6.62
21
14639
947.90
-4.03 +4.15
* 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.
