DeepSeek V4.1 Flash vs Claude Fable 5.1

• DeepSeek V4.1 Flash
• Claude Fable 5.1

Add a model

Add:GPT-6 AstraGemini 4 ArgonMuse Spark 1.1Qwen3.8 Max (0902)

Head to head on 9 boards: Claude Fable 5.1 places best on 8 and DeepSeek V4.1 Flash places best on 1. Both ranked on 9 of 16 boards.

Board	Read	DeepSeek V4.1 Flash	Claude Fable 5.1
Chat1
Text ArenaArena rating1dyesterday	Nov 91dyesterday	#321,475	#61,504best of the picked
Reasoning6
LiveBenchGlobal average1dyesterday	Nov 91dyesterday	#881.1%	#183.4%best of the picked
Epoch ECIECI1dyesterday	Nov 91dyesterday	#19159.3	#5166.9best of the picked
ARC-AGI-2Score1dyesterday	Nov 91dyesterday	#1978.1%	#790.0%best of the picked
AA IntelligenceIndex1dyesterday	Nov 91dyesterday	#1632.6	#950.0best of the picked
HLEAccuracy1dyesterday	Nov 91dyesterday		#246.5%
GPQA DiamondAccuracy1dyesterday	Nov 91dyesterday	#2193.2%
Coding2
WebDev ArenaArena rating1dyesterday	Nov 91dyesterday	#141,665	#51,746best of the picked
SWE-Bench ProResolved1dyesterday	Nov 91dyesterday		#292.2%
Agents and tools2
MCP AtlasPass rate1dyesterday	Nov 91dyesterday		#387.2%
Search ArenaArena rating1dyesterday	Nov 91dyesterday		#71,227
Vision2
Vision ArenaArena rating1dyesterday	Nov 91dyesterday	#341,263	#131,288best of the picked
Document ArenaArena rating1dyesterday	Nov 91dyesterday		#21,499
Speed1
AA SpeedOutput speed1dyesterday	Nov 91dyesterday	#2514 tok/s	#1422 tok/sbest of the picked
Price1
AA PriceBlended price1dyesterday	Nov 91dyesterday	#3$0.53best of the picked	#17$20.0
Usage1
OpenRouterWeekly tokens1dyesterday	Nov 91dyesterday	#156.0T tok

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.