Skip to content

OpenAI vs DeepSeek vs Microsoft vs MiniMax

Each lab shows its best-ranked model on every board. One row per board where at least one is ranked, from each board's current capture.

  • OpenAI
  • DeepSeek
  • Microsoft
  • MiniMax

Head to head on 22 boards: OpenAI places best on 16, MiniMax places best on 3, DeepSeek places best on 2 and Microsoft places best on 1. All 4 ranked on 1 of 28 boards.

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

BoardOpenAIDeepSeekMicrosoftMiniMax
Chat2
Text ArenaArena ratingtoday#161,483best of the pickedGPT-5.6 Sol#251,477DeepSeek V4.1 Flash#761,440MiniMax-M3
MultiChallengeScoretoday#469.2%best of the pickedGPT-5.4 Pro#2446.1%DeepSeek V3P1
Reasoning7
MMLU-ProAccuracy · staletoday#389.3%best of the pickedo1#3385.0%DeepSeek-V3.2#7576.0%Phi 4 Plus#688.0%MiniMax M2.1
GPQA DiamondAccuracytoday#195.8%best of the pickedGPT-6 Astra#2091.7%DeepSeek V4 Pro 0813#2590.9%MiniMax-M3
Epoch ECIECItoday#1166.6best of the pickedGPT-6 Astra#24155.4DeepSeek V4 Pro 0813#60147.0MiniMax-M3
ARC-AGI-2Scoretoday#195.0%best of the pickedGPT-6 Astra#2361.4%DeepSeek V4 Flash Vision#524.9%MiniMax-M2.5
AA IntelligenceIndextoday#352.7best of the pickedGPT-6 Astra#1839.5DeepSeek V4.1 Flash#2829.2MiniMax-M3
LiveBenchGlobal averagetoday#482.2%best of the pickedGPT-6 Astra#681.1%DeepSeek V4.1 Flash#5567.3%MiniMax-M3
HLEAccuracytoday#154.8%GPT-6 Astra
Coding4
WebDev ArenaArena ratingtoday#21,792best of the pickedGPT-6 Astra#131,621DeepSeek V4.1 Flash#461,483MiniMax-M3
SWE-bench VerifiedResolved · staletoday#672.8%GPT-5.2#1170.0%DeepSeek-V3.2#275.8%best of the pickedMiniMax-M2.5
Aider PolyglotPass rate · staletoday#188.0%best of the pickedGPT-5#874.2%DeepSeek-V3.2
SWE-Bench ProResolvedtoday#390.2%GPT-6 Astra
Agents and tools3
BFCLOverall accuracytoday#763.0%best of the pickedo3#1256.7%DeepSeek-V3.2#5028.8%Phi 4
Search ArenaArena ratingtoday#11,257GPT-5.6 Sol
MCP AtlasPass ratetoday#1181.8%GPT-5.6
Vision2
Document ArenaArena ratingtoday#61,486best of the pickedGPT-5.5#271,435MiniMax-M3
Vision ArenaArena ratingtoday#111,287best of the pickedGPT-5.5#471,237MiniMax-M3
Image2
Text-to-ImageArena ratingtoday#11,424best of the pickedGPT Image 2.5 Sunburst#41,335MAI Image 2.6
Image EditArena ratingtoday#11,526best of the pickedGPT Image 2.5 Sunburst#51,429MAI Image 2.6
Video2
Text-to-VideoArena ratingtoday#111,368Sora 2 Pro#81,460best of the pickedMiniMax H3
Image-to-VideoArena ratingtoday#11,495MiniMax H3
Speech2
Speech ArenaArena ratingtoday#321,104TTS 1 Hd#261,126Azure Hd 2.5#161,172best of the pickedSpeech 2.8 Hd
Open ASRAverage WERtoday#535.78 WERwhisper-large-v3 (openai)#23.81 WERbest of the pickedazure-speech-07-2026 (microsoft)
Embeddings1
MTEBMean task scoretoday#174.3%harrier-oss-v1-27b (microsoft)
Speed1
AA SpeedOutput speedtoday#5154 tok/sGPT-6 Luna#3232 tok/sbest of the pickedDeepSeek V4.1 Flash
Price1
AA PriceBlended pricetoday#1$0.20best of the pickedGPT-6 Luna#4$0.53DeepSeek V4.1 Flash
Usage1
OpenRouterWeekly tokenstoday#58.6T tokGPT-5.6 Luna#118.6T tokbest of the pickedDeepSeek V4.1 Flash

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

New #1s by email

Saturdays, only in weeks when a leaderboard has a new #1.

Double opt-in. Unsubscribe any time.