AI ML
Post #7435, on Nov 18, 2025 in TG
Gemini 3 Pro Dominates Text Leaderboard While Claude and GPT Battle for Midfield
Description
A screenshot of the LMSYS Chatbot Arena 'Text' leaderboard updated '1 day ago'. Gemini-3-pro leads at #1 with score 1501 (3,089 votes), followed by grok-4.1-thinking at #2 with 1484 (3,394 votes) and grok-4.1 at #3 with 1465 (3,508 votes). Gemini-2.5-pro sits at #4 with 1451 (65,549 votes). Claude models cluster mid-range: claude-sonnet-4-5-20250929-t at #5 (1449, 16,595 votes), claude-opus-4-1-20250805-thi at #6 (1449, 32,125 votes), claude-sonnet-4-5-20250929 at #7 (1445, 11,156 votes). GPT-4.5-preview is at #8 (1442, 14,644 votes) and claude-opus-4-1-20250805 at #9 (1440, 44,844 votes). The top 3 models all have very low vote counts (3-3.5K) compared to established models with 40-65K votes, raising questions about ranking stability
Related Memes
Use J and K for navigation
Comments
2Comment deleted
The top 3 models have fewer votes than a typical Stack Overflow question about centering a div, but sure, let's crown them the best LLMs ever made
The only thing that changes faster than this AI leaderboard is the project requirements three days before a major release