OpenAI vs DeepSeek vs Microsoft vs Anthropic

Head to head on 25 boards: OpenAI places best on 11, Anthropic places best on 10, DeepSeek places best on 2 and Microsoft places best on 2. All 4 ranked on 2 of 27 boards.

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

OpenAIDeepSeekMicrosoftAnthropic

Board	Read	OpenAI	DeepSeek	Microsoft	Anthropic
Chat2
Text ArenaArena ratingtodaytoday	Sep 27todaytoday	#161,483GPT-5.6 Sol	#251,477DeepSeek V4.1 Flash		#11,509best of the pickedClaude Opus 5.5
MultiChallengeScoretodaytoday	Sep 27todaytoday	#469.2%best of the pickedGPT-5.4 Pro	#2446.1%DeepSeek V3P1		#1259.0%Claude Opus 4.5
Reasoning7
MMLU-ProAccuracy · staletodaytoday	Sep 27todaytoday	#389.3%best of the pickedo1	#3385.0%DeepSeek-V3.2	#7576.0%Phi 4 Plus	#489.1%Claude Opus 4.6
AA IntelligenceIndextodaytoday	Sep 27todaytoday	#352.7GPT-6 Astra	#1839.5DeepSeek V4.1 Flash		#157.6best of the pickedClaude Opus 5.5
GPQA DiamondAccuracytodaytoday	Sep 27todaytoday	#195.8%best of the pickedGPT-6 Astra	#2091.7%DeepSeek V4 Pro 0813		#1093.9%Claude Opus 5
Epoch ECIECItodaytoday	Sep 27todaytoday	#1166.6best of the pickedGPT-6 Astra	#24155.4DeepSeek V4 Pro 0813		#2165.0Claude Fable 5.1
ARC-AGI-2Scoretodaytoday	Sep 27todaytoday	#195.0%best of the pickedGPT-6 Astra	#2361.4%DeepSeek V4 Flash Vision		#293.3%Claude Opus 5.5
LiveBenchGlobal averagetodaytoday	Sep 27todaytoday	#482.2%GPT-6 Astra	#681.1%DeepSeek V4.1 Flash		#183.4%best of the pickedClaude Fable 5.1
HLEAccuracytodaytoday	Sep 27todaytoday	#154.8%best of the pickedGPT-6 Astra			#246.5%Claude Fable 5.1
Coding4
WebDev ArenaArena ratingtodaytoday	Sep 27todaytoday	#21,792GPT-6 Astra	#131,621DeepSeek V4.1 Flash		#11,827best of the pickedClaude Opus 5.5
SWE-bench VerifiedResolved · staletodaytoday	Sep 27todaytoday	#672.8%GPT-5.2	#1170.0%DeepSeek-V3.2		#176.8%best of the pickedClaude Opus 4.5
Aider PolyglotPass rate · staletodaytoday	Sep 27todaytoday	#188.0%best of the pickedGPT-5	#874.2%DeepSeek-V3.2		#1072.0%Claude Opus 4
SWE-Bench ProResolvedtodaytoday	Sep 27todaytoday	#390.2%GPT-6 Astra			#198.0%best of the pickedClaude Opus 5
Agents and tools3
BFCLOverall accuracytodaytoday	Sep 27todaytoday	#763.0%o3	#1256.7%DeepSeek-V3.2	#5028.8%Phi 4	#177.5%best of the pickedClaude Opus 4.5
Search ArenaArena ratingtodaytoday	Sep 27todaytoday	#11,257best of the pickedGPT-5.6 Sol			#21,253Claude Opus 4.6
MCP AtlasPass ratetodaytoday	Sep 27todaytoday	#1181.8%GPT-5.6			#287.2%best of the pickedClaude Fable 5.1
Vision2
Vision ArenaArena ratingtodaytoday	Sep 27todaytoday	#111,287GPT-5.5			#11,310best of the pickedClaude Fable 5
Document ArenaArena ratingtodaytoday	Sep 27todaytoday	#61,486GPT-5.5			#11,516best of the pickedClaude Opus 5
Image2
Text-to-ImageArena ratingtodaytoday	Sep 27todaytoday	#11,424best of the pickedGPT Image 2.5 Sunburst		#41,335MAI Image 2.6
Image EditArena ratingtodaytoday	Sep 27todaytoday	#11,526best of the pickedGPT Image 2.5 Sunburst		#51,429MAI Image 2.6
Video1
Text-to-VideoArena ratingtodaytoday	Sep 27todaytoday	#111,368Sora 2 Pro
Speech2
Open ASRAverage WERtodaytoday	Sep 27todaytoday	#535.78 WERwhisper-large-v3 (openai)		#23.81 WERbest of the pickedazure-speech-07-2026 (microsoft)
Speech ArenaArena ratingtodaytoday	Sep 27todaytoday	#321,104TTS 1 Hd		#261,126best of the pickedAzure Hd 2.5
Embeddings1
MTEBMean task scoretodaytoday	Sep 27todaytoday			#174.3%harrier-oss-v1-27b (microsoft)
Speed1
AA SpeedOutput speedtodaytoday	Sep 27todaytoday	#5154 tok/sGPT-6 Luna	#3232 tok/sbest of the pickedDeepSeek V4.1 Flash		#988 tok/sClaude Opus 5.5
Price1
AA PriceBlended pricetodaytoday	Sep 27todaytoday	#1$0.20best of the pickedGPT-6 Luna	#4$0.53DeepSeek V4.1 Flash		#17$4.00Claude Sonnet 5
Usage1
OpenRouterWeekly tokenstodaytoday	Sep 27todaytoday	#58.6T tokGPT-5.6 Luna	#118.6T tokbest of the pickedDeepSeek V4.1 Flash		#191.5T tokClaude Sonnet 5

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.