Current Rankings
Updated 4:55:45 PMAuto-refresh · 30s
Provisional: some items have overlapping 95% intervals, so they share a rank until more votes separate them.
=1
đ
Qwen3.5-397B
8 matches88% win ratecould rank #1â5
1882
±309 95%
=1
đ
DeepSeek V3
12 matches58% win ratecould rank #1â5
1691
±244 95%
=1
đ
Llama 3.3 70B
12 matches42% win ratecould rank #1â5
1587
±252 95%
=1
đ
Gemma 2 27B
8 matches25% win ratecould rank #1â5
1509
±306 95%
=1
đ
Mistral Large
4 matches25% win ratecould rank #1â5
1344
±425 95%
About This Ranking
DEMO DATA - votes here were seeded to show the format, not cast by real people. Vote to make it real.
5
Items
22
Votes
Help Improve Rankings
Your votes help calibrate the rankings. Each comparison makes them more accurate!
Start VotingShare Ranking
Scan to vote