Gemini 3.1 Pro Preview vs Claude Fable 5.1 vs GPT-6 Astra

One row per board where at least one is ranked, from each board's current capture.

Gemini 3.1 Pro Preview
Claude Fable 5.1
GPT-6 Astra

Add:

Muse Spark 1.1

Qwen3.8 Max (0902)Inkling Small 2

DeepSeek V4P1 Flash

Head to head on 15 boards: Claude Fable 5.1 places best on 8, GPT-6 Astra places best on 8 and Gemini 3.1 Pro Preview places best on 0 (a tie counts for each). All 3 ranked on 9 of 17 boards.

Two at a time on this screen. Tap one to swap it in.

Board	Read	Gemini 3.1 Pro Preview	Claude Fable 5.1	GPT-6 Astra
Chat2
Text ArenaArena ratingyesterday	Nov 9yesterday	#171,487	#61,504best of the picked	#311,475
MultiChallengeScoreyesterday	Nov 9yesterday	#871.4%
Reasoning7
HLEAccuracyyesterday	Nov 9yesterday	#346.4%	#246.5%	#154.8%best of the picked
Epoch ECIECIyesterday	Nov 9yesterday	#34156.2	#5166.9	#1169.1best of the picked
ARC-AGI-2Scoreyesterday	Nov 9yesterday	#2077.1%	#790.0%	#195.0%best of the picked
LiveBenchGlobal averageyesterday	Nov 9yesterday	#2577.0%	#183.4%best of the picked	#482.2%
GPQA DiamondAccuracyyesterday	Nov 9yesterday	#894.4%		#295.8%best of the picked
AA IntelligenceIndexyesterday	Nov 9yesterday		#950.0	#356.9best of the picked
MMLU-ProAccuracy · staleyesterday	Nov 9yesterday	#191.2%
Coding2
WebDev ArenaArena ratingyesterday	Nov 9yesterday	#671,447	#51,746	#31,770best of the picked
SWE-Bench ProResolvedyesterday	Nov 9yesterday		#292.2%best of the picked	#490.2%
Agents and tools2
Search ArenaArena ratingyesterday	Nov 9yesterday	#141,208	#71,227	#31,245best of the picked
MCP AtlasPass rateyesterday	Nov 9yesterday	#1778.2%	#387.2%best of the picked
Vision2
Document ArenaArena ratingyesterday	Nov 9yesterday	#231,459	#21,499best of the picked	#111,477
Vision ArenaArena ratingyesterday	Nov 9yesterday	#221,280	#131,288best of the picked	#171,285
Speed1
AA SpeedOutput speedyesterday	Nov 9yesterday		#1422 tok/sbest of the picked	#1621 tok/s
Price1
AA PriceBlended priceyesterday	Nov 9yesterday		#17$20.0best of the picked	#17$20.0best 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.

More comparisons