Every AI leaderboard

28 leaderboards · 992 models ranked

Sprag vs Anthropic vs Meta vs DeepSeek

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.

  • Sprag
  • Anthropic
  • Meta
  • DeepSeek

Head to head on 21 boards: Anthropic places best on 15, Meta places best on 4, DeepSeek places best on 2 and Sprag places best on 1 (a tie counts for each). All 4 ranked on 0 of 26 boards.

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

BoardSpragAnthropicMetaDeepSeek
Chat2
Text ArenaArena ratingyesterday#41,506Claude Opus 5.5#31,511best of the pickedMuse Wonder#321,475DeepSeek V4.1 Flash
MultiChallengeScoreyesterday#1067.7%Claude Haiku 5.5#375.5%best of the pickedMuse Spark#574.9%DeepSeek V4P1 Flash
Reasoning7
AA IntelligenceIndexyesterday#164.6best of the pickedClaude Fable 5.2#950.1Muse Wonder Full#1732.6DeepSeek V4.1 Flash
Epoch ECIECIyesterday#1174.2best of the pickedClaude Fable 5.2#15162.0Muse Wonder#21159.3DeepSeek V4.1 Flash
LiveBenchGlobal averageyesterday#183.4%best of the pickedClaude Fable 5.1#681.6%Muse Spark 1.3#881.1%DeepSeek V4.1 Flash
GPQA DiamondAccuracyyesterday#495.6%best of the pickedClaude Sonnet 5.5#4989.8%Muse Spark#2393.2%DeepSeek V4.1 Flash
ARC-AGI-2Scoreyesterday#493.8%best of the pickedClaude Fable 5.2#4157.5%Muse Spark Open#2178.1%DeepSeek V4.1 Flash
MMLU-ProAccuracy · staleyesterday#489.1%best of the pickedClaude Opus 4.6#4782.8%Llama4 Behemoth#3385.0%DeepSeek-V3.2
HLEAccuracyyesterday#246.5%best of the pickedClaude Fable 5.1#740.6%Muse Spark
Coding4
WebDev ArenaArena ratingyesterday#11,836best of the pickedClaude Fable 5.2#81,723Muse Wonder Full#151,664DeepSeek V4.1 Flash
SWE-bench VerifiedResolved · staleyesterday#176.8%best of the pickedClaude Opus 4.5#3621.0%Llama 4 Maverick Instruct#1170.0%DeepSeek-V3.2
Aider PolyglotPass rate · staleyesterday#1072.0%Claude Opus 4#4315.6%Llama 4 Maverick Instruct#874.2%best of the pickedDeepSeek-V3.2
SWE-Bench ProResolvedyesterday#198.0%best of the pickedClaude Opus 5#490.2%Muse Wonder 1.0 Max Effort
Agents and tools3
BFCLOverall accuracy · staleyesterday#177.5%best of the pickedClaude Opus 4.5#3637.3%Llama 4 Maverick 17b 128e#1256.7%DeepSeek-V3.2
MCP AtlasPass rateyesterday#387.2%Claude Fable 5.1#288.1%best of the pickedMuse Spark 1.1
Search ArenaArena ratingyesterday#51,229Claude Fable 5
Vision2
Vision ArenaArena ratingyesterday#11,309best of the pickedClaude Fable 5#61,294Muse Spark#351,263DeepSeek V4.1 Flash
Document ArenaArena ratingyesterday#11,500best of the pickedClaude Fable 5#91,480Muse Spark 1.1
Image2
Text-to-ImageArena ratingyesterday#111,266Muse Image
Image EditArena ratingyesterday#121,400Muse Image
Video1
Text-to-VideoArena ratingyesterday#101,457Muse Video
Speech2
Open ASRAverage WERyesterday#206.85 WERbest of the pickedsymphony (sprag)#307.00 WERmuse-voice-transcribe (meta)
Speech ArenaArena ratingyesterday#99846Metavoice V1
Speed1
AA SpeedOutput speedyesterday#444 tok/sbest of the pickedClaude Fable 5.2#2513 tok/sMuse Wonder#1524 tok/sDeepSeek V4 Pro 0813
Price1
AA PriceBlended priceyesterday#1$0.20best of the pickedClaude Haiku 5.5#1$0.20best of the pickedMuse Ember#4$0.53DeepSeek V4.1 Flash
Usage1
OpenRouterWeekly tokensyesterday#66.8T tokClaude Opus 5.5#152.9T tokMuse Spark 1.3 Contributor#153.8T 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