Wan 2.5 Gemini 4 Argon Claude Fable 5.1 Head to head on 9 boards: Gemini 4 Argon places best on 6 , Claude Fable 5.1 places best on 3 and Wan 2.5 places best on 0 . All 3 ranked on 0 of 20 boards.
Board Read Wan 2.5 Gemini 4 Argon Claude Fable 5.1 Chat2 Text Arena Arena rating 1d yesterday Nov 9 1d yesterday #2 1,523 best of the picked #6 1,504 MultiChallenge Score 1d yesterday Nov 9 1d yesterday #2 78.7% Reasoning6 Epoch ECI ECI 1d yesterday Nov 9 1d yesterday #6 166.4 #5 166.9 best of the picked ARC-AGI-2 Score 1d yesterday Nov 9 1d yesterday #5 91.3% best of the picked #7 90.0% AA Intelligence Index 1d yesterday Nov 9 1d yesterday #7 50.4 best of the picked #9 50.0 GPQA Diamond Accuracy 1d yesterday Nov 9 1d yesterday #1 96.4% LiveBench Global average 1d yesterday Nov 9 1d yesterday #1 83.4% HLE Accuracy 1d yesterday Nov 9 1d yesterday #2 46.5% Coding2 SWE-Bench Pro Resolved 1d yesterday Nov 9 1d yesterday #3 92.2% #2 92.2% best of the picked WebDev Arena Arena rating 1d yesterday Nov 9 1d yesterday #10 1,683 #5 1,746 best of the picked Agents and tools2 MCP Atlas Pass rate 1d yesterday Nov 9 1d yesterday #1 90.7% best of the picked #3 87.2% Search Arena Arena rating 1d yesterday Nov 9 1d yesterday #7 1,227 Vision2 Document Arena Arena rating 1d yesterday Nov 9 1d yesterday #2 1,499 Vision Arena Arena rating 1d yesterday Nov 9 1d yesterday #13 1,288 Image2 Text-to-Image Arena rating 1d yesterday Nov 9 1d yesterday #41 1,117 Image Edit Arena rating 1d yesterday Nov 9 1d yesterday #45 1,180 Video2 Image-to-Video Arena rating 1d yesterday Nov 9 1d yesterday #21 1,322 Text-to-Video Arena rating 1d yesterday Nov 9 1d yesterday #23 1,244 Speed1 AA Speed Output speed 1d yesterday Nov 9 1d yesterday #10 28 tok/s best of the picked #14 22 tok/s Price1 AA Price Blended price 1d yesterday Nov 9 1d yesterday #13 $8.00 best of the picked #17 $20.0
Green marks the best placing among the picked where two or more are ranked; ties are marked for each. Ranks are each board's own order (lower is better), scores as the board published them. Dash: not on that board.