Claude Haiku 5.5 vs GPT-6 Astra vs Kimi K3 vs Muse Spark

Head to head on 14 boards: GPT-6 Astra places best on 9, Claude Haiku 5.5 places best on 2, Muse Spark places best on 2 and Kimi K3 places best on 1. All 4 ranked on 0 of 16 boards.

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

Claude Haiku 5.5GPT-6 AstraKimi K3Muse Spark

Board	Read	Claude Haiku 5.5	GPT-6 Astra	Kimi K3	Muse Spark
Chat2
Text ArenaArena ratingtodaytoday	Oct 8todaytoday		#261,477	#131,488	#121,489best of the picked
MultiChallengeScoretodaytoday	Oct 8todaytoday				#275.5%
Reasoning6
GPQA DiamondAccuracytodaytoday	Oct 8todaytoday		#195.8%best of the picked	#1893.1%	#4189.8%
Epoch ECIECItodaytoday	Oct 8todaytoday		#2166.4best of the picked	#15157.4	#44152.0
AA IntelligenceIndextodaytoday	Oct 8todaytoday	#1443.4	#452.7best of the picked	#1343.6
LiveBenchGlobal averagetodaytoday	Oct 8todaytoday	#4672.1%	#482.2%best of the picked	#1379.2%
HLEAccuracytodaytoday	Oct 8todaytoday		#154.8%best of the picked		#640.6%
ARC-AGI-2Scoretodaytoday	Oct 8todaytoday		#195.0%best of the picked	#3160.4%
Coding2
WebDev ArenaArena ratingtodaytoday	Oct 8todaytoday		#21,788best of the picked	#111,655
SWE-Bench ProResolvedtodaytoday	Oct 8todaytoday		#390.2%best of the picked	#488.2%
Agents and tools1
MCP AtlasPass ratetodaytoday	Oct 8todaytoday			#882.3%best of the picked	#982.2%
Vision2
Vision ArenaArena ratingtodaytoday	Oct 8todaytoday		#181,284		#61,294best of the picked
Document ArenaArena ratingtodaytoday	Oct 8todaytoday		#131,468best of the picked		#241,444
Speed1
AA SpeedOutput speedtodaytoday	Oct 8todaytoday	#2243 tok/sbest of the picked	#2347 tok/s	#2542 tok/s
Price1
AA PriceBlended pricetodaytoday	Oct 8todaytoday	#2$0.20best of the picked	#27$20.0	#25$6.00
Usage1
OpenRouterWeekly tokenstodaytoday	Oct 8todaytoday			#151.7T 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.