Wan 2.5 Claude Fable 5.1 GPT-6 Astra Head to head on 13 boards: Claude Fable 5.1 places best on 7 , GPT-6 Astra places best on 7 and Wan 2.5 places best on 0 (a tie counts for each). All 3 ranked on 0 of 19 boards.
Board Read Wan 2.5 Claude Fable 5.1 GPT-6 Astra Chat1 Text Arena Arena rating today today Nov 9 today today #6 1,504 best of the picked #31 1,475 Reasoning6 HLE Accuracy today today Nov 9 today today #2 46.5% #1 54.8% best of the picked Epoch ECI ECI today today Nov 9 today today #5 166.9 #1 169.1 best of the picked ARC-AGI-2 Score today today Nov 9 today today #7 90.0% #1 95.0% best of the picked LiveBench Global average today today Nov 9 today today #1 83.4% best of the picked #4 82.2% AA Intelligence Index today today Nov 9 today today #9 50.0 #3 56.9 best of the picked GPQA Diamond Accuracy today today Nov 9 today today #2 95.8% Coding2 SWE-Bench Pro Resolved today today Nov 9 today today #2 92.2% best of the picked #4 90.2% WebDev Arena Arena rating today today Nov 9 today today #5 1,746 #3 1,770 best of the picked Agents and tools2 Search Arena Arena rating today today Nov 9 today today #7 1,227 #3 1,245 best of the picked MCP Atlas Pass rate today today Nov 9 today today #3 87.2% Vision2 Document Arena Arena rating today today Nov 9 today today #2 1,499 best of the picked #11 1,477 Vision Arena Arena rating today today Nov 9 today today #13 1,288 best of the picked #17 1,285 Image2 Text-to-Image Arena rating today today Nov 9 today today #41 1,117 Image Edit Arena rating today today Nov 9 today today #45 1,180 Video2 Image-to-Video Arena rating today today Nov 9 today today #21 1,322 Text-to-Video Arena rating today today Nov 9 today today #23 1,244 Speed1 AA Speed Output speed today today Nov 9 today today #14 22 tok/s best of the picked #16 21 tok/s Price1 AA Price Blended price today today Nov 9 today today #17 $20.0 best of the picked #17 $20.0 best of the picked
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.