Claude Opus 4 vs Gemini 4 Argon vs Inkling Small 2 vs GPT-6 Astra

Head to head on 14 boards: Gemini 4 Argon places best on 7, GPT-6 Astra places best on 7, Claude Opus 4 places best on 0 and Inkling Small 2 places best on 0. All 4 ranked on 0 of 19 boards.

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

Claude Opus 4Gemini 4 Argon?Inkling Small 2GPT-6 Astra

Board	Read	Claude Opus 4	Gemini 4 Argon	?Inkling Small 2	GPT-6 Astra
Chat2
Text ArenaArena ratingtodaytoday	Nov 9todaytoday		#21,523best of the picked		#311,475
MultiChallengeScoretodaytoday	Nov 9todaytoday	#1958.6%	#278.7%best of the picked
Reasoning7
Epoch ECIECItodaytoday	Nov 9todaytoday	#98142.8	#6166.4		#1169.1best of the picked
ARC-AGI-2Scoretodaytoday	Nov 9todaytoday	#598.6%	#591.3%		#195.0%best of the picked
HLEAccuracytodaytoday	Nov 9todaytoday	#2610.7%			#154.8%best of the picked
GPQA DiamondAccuracytodaytoday	Nov 9todaytoday		#196.4%best of the picked		#295.8%
AA IntelligenceIndextodaytoday	Nov 9todaytoday		#750.4		#356.9best of the picked
LiveBenchGlobal averagetodaytoday	Nov 9todaytoday				#482.2%
MMLU-ProAccuracy · staletodaytoday	Nov 9todaytoday	#1287.3%
Coding4
SWE-Bench ProResolvedtodaytoday	Nov 9todaytoday		#392.2%best of the picked	#986.3%	#490.2%
WebDev ArenaArena ratingtodaytoday	Nov 9todaytoday		#101,683		#31,770best of the picked
Aider PolyglotPass rate · staletodaytoday	Nov 9todaytoday	#1072.0%
SWE-bench VerifiedResolved · staletodaytoday	Nov 9todaytoday	#1267.6%
Agents and tools2
MCP AtlasPass ratetodaytoday	Nov 9todaytoday		#190.7%best of the picked	#585.4%
Search ArenaArena ratingtodaytoday	Nov 9todaytoday	#351,152			#31,245best of the picked
Vision2
Vision ArenaArena ratingtodaytoday	Nov 9todaytoday	#741,208			#171,285best of the picked
Document ArenaArena ratingtodaytoday	Nov 9todaytoday				#111,477
Speed1
AA SpeedOutput speedtodaytoday	Nov 9todaytoday		#1028 tok/sbest of the picked		#1621 tok/s
Price1
AA PriceBlended pricetodaytoday	Nov 9todaytoday		#13$8.00best 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.