Skip to content

Google vs DeepSeek vs Anthropic vs OpenAI

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.

  • Google
  • DeepSeek
  • Anthropic
  • OpenAI

Head to head on 26 boards: Anthropic places best on 10, OpenAI places best on 10, Google places best on 5 and DeepSeek places best on 1. All 4 ranked on 15 of 28 boards.

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

BoardGoogleDeepSeekAnthropicOpenAI
Chat2
Text ArenaArena ratingtoday#81,492Gemini 3.8 Flash#251,477DeepSeek V4.1 Flash#11,509best of the pickedClaude Opus 5.5#161,483GPT-5.6 Sol
MultiChallengeScoretoday#371.4%best of the pickedGemini 3.1 Pro Preview#2446.1%DeepSeek V3P1#1259.0%Claude Opus 4.5#469.2%GPT-5.4 Pro
Reasoning7
AA IntelligenceIndextoday#1540.9Gemini 3.8 Flash#1839.5DeepSeek V4.1 Flash#157.6best of the pickedClaude Opus 5.5#352.7GPT-6 Astra
GPQA DiamondAccuracytoday#295.4%Gemini 3.8 Flash#2091.7%DeepSeek V4 Pro 0813#1093.9%Claude Opus 5#195.8%best of the pickedGPT-6 Astra
Epoch ECIECItoday#11157.7Gemini 3.7 Flash#24155.4DeepSeek V4 Pro 0813#2165.0Claude Fable 5.1#1166.6best of the pickedGPT-6 Astra
ARC-AGI-2Scoretoday#689.2%Gemini 3.8 Flash#2361.4%DeepSeek V4 Flash Vision#293.3%Claude Opus 5.5#195.0%best of the pickedGPT-6 Astra
LiveBenchGlobal averagetoday#1378.8%Gemini 3.7 Flash#681.1%DeepSeek V4.1 Flash#183.4%best of the pickedClaude Fable 5.1#482.2%GPT-6 Astra
MMLU-ProAccuracy · staletoday#191.2%best of the pickedGemini 3.1 Pro Preview#3385.0%DeepSeek-V3.2#489.1%Claude Opus 4.6#389.3%o1
HLEAccuracytoday#346.4%Gemini 3.1 Pro Preview#246.5%Claude Fable 5.1#154.8%best of the pickedGPT-6 Astra
Coding4
WebDev ArenaArena ratingtoday#201,593Gemini 3.7 Flash#131,621DeepSeek V4.1 Flash#11,827best of the pickedClaude Opus 5.5#21,792GPT-6 Astra
SWE-bench VerifiedResolved · staletoday#275.8%Gemini 3 Flash#1170.0%DeepSeek-V3.2#176.8%best of the pickedClaude Opus 4.5#672.8%GPT-5.2
Aider PolyglotPass rate · staletoday#383.1%Gemini 2.5 Pro (Jun 2025)#874.2%DeepSeek-V3.2#1072.0%Claude Opus 4#188.0%best of the pickedGPT-5
SWE-Bench ProResolvedtoday#958.8%Gemini 3.8 Flash#198.0%best of the pickedClaude Opus 5#390.2%GPT-6 Astra
Agents and tools3
BFCLOverall accuracytoday#372.5%Gemini 3 Pro#1256.7%DeepSeek-V3.2#177.5%best of the pickedClaude Opus 4.5#763.0%o3
Search ArenaArena ratingtoday#91,210Gemini 3.1 Pro Grounding#21,253Claude Opus 4.6#11,257best of the pickedGPT-5.6 Sol
MCP AtlasPass ratetoday#683.6%Gemini 3.5 Flash#287.2%best of the pickedClaude Fable 5.1#1181.8%GPT-5.6
Vision2
Vision ArenaArena ratingtoday#101,289Gemini 3 Pro#11,310best of the pickedClaude Fable 5#111,287GPT-5.5
Document ArenaArena ratingtoday#161,463Gemini 3.5 Flash#11,516best of the pickedClaude Opus 5#61,486GPT-5.5
Image2
Text-to-ImageArena ratingtoday#91,261Gemini 3.1 Flash Image#11,424best of the pickedGPT Image 2.5 Sunburst
Image EditArena ratingtoday#91,390Gemini 3 Pro Image#11,526best of the pickedGPT Image 2.5 Sunburst
Video2
Text-to-VideoArena ratingtoday#11,516best of the pickedGemini Omni 1.1 Flash#111,368Sora 2 Pro
Image-to-VideoArena ratingtoday#21,488Gemini Omni 1.1 Flash
Speech2
Speech ArenaArena ratingtoday#21,268best of the pickedGemini 3.8 Flash TTS#321,104TTS 1 Hd
Open ASRAverage WERtoday#669.51 WERgemma-4-E4B-it (google)#535.78 WERbest of the pickedwhisper-large-v3 (openai)
Embeddings1
MTEBMean task scoretoday#568.4%gemini-embedding-001 (google)
Speed1
AA SpeedOutput speedtoday#1308 tok/sbest of the pickedGemini 3.8 Flash#3232 tok/sDeepSeek V4.1 Flash#988 tok/sClaude Opus 5.5#5154 tok/sGPT-6 Luna
Price1
AA PriceBlended pricetoday#9$1.50Gemini 3.8 Flash#4$0.53DeepSeek V4.1 Flash#17$4.00Claude Sonnet 5#1$0.20best of the pickedGPT-6 Luna
Usage1
OpenRouterWeekly tokenstoday#142.2T tokGemini 3.8 Flash#118.6T tokbest of the pickedDeepSeek V4.1 Flash#191.5T tokClaude Sonnet 5#58.6T tokGPT-5.6 Luna

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.