Every AI leaderboard

28 leaderboards · 915 models ranked

P99lab vs Google vs DeepSeek vs Meta

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.

  • P99lab
  • Google
  • DeepSeek
  • Meta

Head to head on 25 boards: Google places best on 16, Meta places best on 6, DeepSeek places best on 2 and P99lab places best on 1. All 4 ranked on 0 of 28 boards.

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

BoardP99labGoogleDeepSeekMeta
Chat2
Text ArenaArena ratingyesterday#11,517best of the pickedGemini 4 Argon#291,476DeepSeek V4.1 Flash#31,509Muse Wonder
MultiChallengeScoreyesterday#278.7%best of the pickedGemini 4 Argon#2846.1%DeepSeek V3P1#375.5%Muse Spark
Reasoning7
GPQA DiamondAccuracyyesterday#196.4%best of the pickedGemini 4 Argon#1993.2%DeepSeek V4.1 Flash#4389.8%Muse Spark
MMLU-ProAccuracy · staleyesterday#191.2%best of the pickedGemini 3.1 Pro Preview#3385.0%DeepSeek-V3.2#4782.8%Llama4 Behemoth
AA IntelligenceIndexyesterday#752.6Gemini 4 Argon#1445.1DeepSeek V4.1 Flash#355.8best of the pickedMuse Wonder Full
Epoch ECIECIyesterday#5166.4best of the pickedGemini 4 Argon#32155.9DeepSeek V4.1 Flash#12161.8Muse Wonder
ARC-AGI-2Scoreyesterday#591.3%best of the pickedGemini 4 Argon#1878.1%DeepSeek V4.1 Flash#3857.5%Muse Spark Open
LiveBenchGlobal averageyesterday#1578.8%Gemini 3.7 Flash#881.1%DeepSeek V4.1 Flash#681.6%best of the pickedMuse Spark 1.3
HLEAccuracyyesterday#346.4%best of the pickedGemini 3.1 Pro Preview#740.6%Muse Spark
Coding4
SWE-bench VerifiedResolved · staleyesterday#275.8%best of the pickedGemini 3 Flash#1170.0%DeepSeek-V3.2#3621.0%Llama 4 Maverick Instruct
Aider PolyglotPass rate · staleyesterday#383.1%best of the pickedGemini 2.5 Pro (Jun 2025)#874.2%DeepSeek-V3.2#4315.6%Llama 4 Maverick Instruct
WebDev ArenaArena ratingyesterday#101,674Gemini 4 Argon#121,665DeepSeek V4.1 Flash#61,726best of the pickedMuse Wonder Full
SWE-Bench ProResolvedyesterday#392.2%best of the pickedGemini 4 Argon#490.2%Muse Wonder 1.0 Max Effort
Agents and tools3
BFCLOverall accuracy · staleyesterday#372.5%best of the pickedGemini 3 Pro#1256.7%DeepSeek-V3.2#3637.3%Llama 4 Maverick 17b 128e
MCP AtlasPass rateyesterday#683.6%Gemini 3.5 Flash#188.1%best of the pickedMuse Spark 1.1
Search ArenaArena ratingyesterday#141,207Gemini 3.1 Pro Grounding
Vision2
Vision ArenaArena ratingyesterday#51,295best of the pickedGemini 3.7 Flash#391,261DeepSeek V4.1 Flash#61,294Muse Spark
Document ArenaArena ratingyesterday#211,461Gemini 3.5 Flash#91,479best of the pickedMuse Spark 1.1
Image2
Image EditArena ratingyesterday#51,426best of the pickedGemini Nano Banana 2.1#91,398Muse Image
Text-to-ImageArena ratingyesterday#61,328best of the pickedGemini Nano Banana 2.1#81,274Muse Image
Video2
Text-to-VideoArena ratingyesterday#11,516best of the pickedGemini Omni 1.1 Flash#91,456Muse Video
Image-to-VideoArena ratingyesterday#21,488Gemini Omni 1.1 Flash
Speech2
Open ASRAverage WERyesterday#34.98 WERbest of the pickedaudrey-0.7 (p99lab)#5611.86 WERgemma-4-E4B-it (google)#326.30 WERmuse-voice-transcribe (meta)
Speech ArenaArena ratingyesterday#51,280best of the pickedGemini 3.8 Flash TTS#95846Metavoice V1
Embeddings1
MTEBMean task scoreyesterday#568.4%gemini-embedding-001 (google)
Speed1
AA SpeedOutput speedyesterday#11132 tok/sGemini 3.8 Flash#8181 tok/sDeepSeek V4 Flash Vision#6195 tok/sbest of the pickedMuse Wonder
Price1
AA PriceBlended priceyesterday#15$1.50Gemini 3.8 Flash#9$0.53best of the pickedDeepSeek V4.1 Flash#17$2.00Muse Spark Open
Usage1
OpenRouterWeekly tokensyesterday#133.6T tokbest of the pickedDeepSeek V4.1 Flash#152.8T 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.

More comparisons