Reflection vs Anthropic vs OpenAI vs Meta

Head to head on 26 boards: Anthropic places best on 12, OpenAI places best on 12, Meta places best on 2 and Reflection places best on 1 (a tie counts for each). All 4 ranked on 2 of 26 boards.

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

RReflectionAnthropicOpenAIMeta

Board	Read	RReflection	Anthropic	OpenAI	Meta
Chat2
MultiChallengeScoretodaytoday	Oct 21todaytoday		#667.7%Claude Haiku 5.5	#182.0%best of the pickedGPT-6.1 Sol	#275.5%Muse Spark
Text ArenaArena ratingtodaytoday	Oct 21todaytoday		#21,507best of the pickedClaude Opus 5.5	#171,485GPT-5.6 Sol	#81,494Muse Spark 1.3
Reasoning7
ARC-AGI-2Scoretodaytoday	Oct 21todaytoday	#4623.5%Beam 1.0	#393.3%Claude Opus 5.5	#195.0%best of the pickedGPT-6 Astra	#800.0%Llama 4 Maverick Instruct
AA IntelligenceIndextodaytoday	Oct 21todaytoday		#158.0best of the pickedClaude Opus 5.5	#453.0GPT-6 Astra	#256.0Muse Wonder Full
HLEAccuracytodaytoday	Oct 21todaytoday		#246.5%Claude Fable 5.1	#154.8%best of the pickedGPT-6 Astra	#740.6%Muse Spark
Epoch ECIECItodaytoday	Oct 21todaytoday		#1168.8best of the pickedClaude Opus 5.5	#2168.8GPT-6 Astra	#12161.8Muse Wonder
LiveBenchGlobal averagetodaytoday	Oct 21todaytoday		#183.4%best of the pickedClaude Fable 5.1	#482.2%GPT-6 Astra	#681.6%Muse Spark 1.3
GPQA DiamondAccuracytodaytoday	Oct 21todaytoday		#395.6%Claude Sonnet 5.5	#295.8%best of the pickedGPT-6 Astra	#4389.8%Muse Spark
MMLU-ProAccuracy · staletodaytoday	Oct 21todaytoday		#489.1%Claude Opus 4.6	#389.3%best of the pickedo1	#4782.8%Llama4 Behemoth
Coding4
WebDev ArenaArena ratingtodaytoday	Oct 21todaytoday		#11,814best of the pickedClaude Opus 5.5	#21,788GPT-6 Astra	#101,656Muse Spark 1.3
SWE-Bench ProResolvedtodaytoday	Oct 21todaytoday		#198.0%best of the pickedClaude Opus 5	#490.2%GPT-6 Astra	#490.2%Muse Wonder 1.0 Max Effort
SWE-bench VerifiedResolved · staletodaytoday	Oct 21todaytoday		#176.8%best of the pickedClaude Opus 4.5	#672.8%GPT-5.2	#3621.0%Llama 4 Maverick Instruct
Aider PolyglotPass rate · staletodaytoday	Oct 21todaytoday		#1072.0%Claude Opus 4	#188.0%best of the pickedGPT-5	#4315.6%Llama 4 Maverick Instruct
Agents and tools3
BFCLOverall accuracy · staletodaytoday	Oct 21todaytoday		#177.5%best of the pickedClaude Opus 4.5	#763.0%o3	#3637.3%Llama 4 Maverick 17b 128e
MCP AtlasPass ratetodaytoday	Oct 21todaytoday		#287.2%Claude Fable 5.1	#1181.8%GPT-5.6	#188.1%best of the pickedMuse Spark 1.1
Search ArenaArena ratingtodaytoday	Oct 21todaytoday		#21,253Claude Opus 4.6	#11,257best of the pickedGPT-5.6 Sol
Vision2
Vision ArenaArena ratingtodaytoday	Oct 21todaytoday		#11,308best of the pickedClaude Fable 5	#61,294GPT-6.1 Sol	#71,294Muse Spark
Document ArenaArena ratingtodaytoday	Oct 21todaytoday		#11,516best of the pickedClaude Opus 5	#61,486GPT-5.5	#121,471Muse Spark 1.3
Image2
Text-to-ImageArena ratingtodaytoday	Oct 21todaytoday			#11,425best of the pickedGPT Image 2.5 Sunburst	#81,274Muse Image
Image EditArena ratingtodaytoday	Oct 21todaytoday			#11,524best of the pickedGPT Image 2.5 Sunburst	#71,403Muse Image
Video1
Text-to-VideoArena ratingtodaytoday	Oct 21todaytoday			#111,368Sora 2 Pro	#91,456best of the pickedMuse Video
Speech2
Open ASRAverage WERtodaytoday	Oct 21todaytoday			#306.38 WERbest of the pickedgpt-transcribe (openai)	#356.48 WERmuse-voice-transcribe (meta)
Speech ArenaArena ratingtodaytoday	Oct 21todaytoday			#391,103best of the pickedTTS 1 Hd	#95847Metavoice V1
Speed1
AA SpeedOutput speedtodaytoday	Oct 21todaytoday	#6345 tok/sbest of the pickedBeam 1.0	#12236 tok/sClaude Haiku 5.5	#9266 tok/sGPT OSS 20b	#7281 tok/sMuse Spark 1.3
Price1
AA PriceBlended pricetodaytoday	Oct 21todaytoday		#3$0.20best of the pickedClaude Haiku 5.5	#3$0.20best of the pickedGPT-6 Luna	#15$2.00Muse Spark 1.3
Usage1
OpenRouterWeekly tokenstodaytoday	Oct 21todaytoday		#56.8T tokbest of the pickedClaude Opus 5.5	#66.2T tokGPT-6 Luna	#123.3T tokMuse Spark 1.3 Contributor

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.