Apex
Showdown
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 Report0 prompts
Real conversation prompts compared across models through pairwise votes.
0 users
From 80+ countries and 70+ languages, spanning all backgrounds and professions.
Style Control
RankModel Votes Score
1gpt-5.3-codex8,508
1145.18-3.49 +5.47
1gemini-3.1-pro8,388
1137.36-4.74 +4.76
2claude-opus-4.6 (Thinking)7,423
1130.06-4.01 +4.50
3gemini-3-deep-think14,115
1128.20-4.08 +3.80
3claude-opus-4.69,560
1126.73-4.10 +4.50
3gpt-5.3-codex-spark10,662
1124.03-3.86 +4.60
6claude-sonnet-4.614,531
1117.52-3.65 +4.30
8claude-sonnet-4.6 (Thinking)14,517
1109.41-4.10 +3.34
8gpt-5-chat11,529
1106.70-4.10 +4.69
8qwen3-235b-a22b-2507-v112,602
1105.72-3.86 +4.79
9gpt-5.1-2025-11-medium9,861
1098.39-4.33 +4.29
10grok-4.207,842
1091.48-5.14 +5.16