Every AI leaderboard

Leaderboards28Models ranked875

28 leaderboards · 875 models ranked

Reflection vs Meta vs DeepSeek vs Google

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.

  • Reflection
  • Meta
  • DeepSeek
  • Google

Head to head on 25 boards: Google places best on 17, Meta places best on 6, DeepSeek places best on 2 and Reflection places best on 0. All 4 ranked on 2 of 28 boards.

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

BoardReflectionMetaDeepSeekGoogle
Chat2
Text ArenaArena ratingtoday#81,494Muse Spark 1.3#341,475DeepSeek V4.1 Flash#11,525best of the pickedGemini 4 Argon
MultiChallengeScoretoday#275.5%best of the pickedMuse Spark#2646.1%DeepSeek V3P1#471.4%Gemini 3.1 Pro Preview
Reasoning7
ARC-AGI-2Scoretoday#4623.5%Beam 1.0#800.0%Llama 4 Maverick Instruct#2172.9%DeepSeek V4.1 Flash#591.3%best of the pickedGemini 4 Argon
GPQA DiamondAccuracytoday#4389.8%Muse Spark#2591.7%DeepSeek V4 Pro 0813#196.4%best of the pickedGemini 4 Argon
MMLU-ProAccuracy · staletoday#4782.8%Llama4 Behemoth#3385.0%DeepSeek-V3.2#191.2%best of the pickedGemini 3.1 Pro Preview
Epoch ECIECItoday#12161.8Muse Wonder#32155.9DeepSeek V4.1 Flash#5166.4best of the pickedGemini 4 Argon
LiveBenchGlobal averagetoday#681.6%best of the pickedMuse Spark 1.3#881.1%DeepSeek V4.1 Flash#1578.8%Gemini 3.7 Flash
AA IntelligenceIndextoday#256.0best of the pickedMuse Wonder Full#453.0Gemini 4 Argon
HLEAccuracytoday#740.6%Muse Spark#346.4%best of the pickedGemini 3.1 Pro Preview
Coding4
SWE-bench VerifiedResolved · staletoday#3621.0%Llama 4 Maverick Instruct#1170.0%DeepSeek-V3.2#275.8%best of the pickedGemini 3 Flash
Aider PolyglotPass rate · staletoday#4315.6%Llama 4 Maverick Instruct#874.2%DeepSeek-V3.2#383.1%best of the pickedGemini 2.5 Pro (Jun 2025)
WebDev ArenaArena ratingtoday#101,656Muse Spark 1.3#191,619DeepSeek V4.1 Flash#81,678best of the pickedGemini 4 Argon
SWE-Bench ProResolvedtoday#490.2%Muse Wonder 1.0 Max Effort#392.2%best of the pickedGemini 4 Argon
Agents and tools3
BFCLOverall accuracy · staletoday#3637.3%Llama 4 Maverick 17b 128e#1256.7%DeepSeek-V3.2#372.5%best of the pickedGemini 3 Pro
MCP AtlasPass ratetoday#188.1%best of the pickedMuse Spark 1.1#683.6%Gemini 3.5 Flash
Search ArenaArena ratingtoday#91,210Gemini 3.1 Pro Grounding
Vision2
Vision ArenaArena ratingtoday#71,294Muse Spark#391,260DeepSeek V4.1 Flash#51,297best of the pickedGemini 3.7 Flash
Document ArenaArena ratingtoday#121,471best of the pickedMuse Spark 1.3#161,463Gemini 3.5 Flash
Image2
Text-to-ImageArena ratingtoday#81,274Muse Image#61,328best of the pickedGemini Nano Banana 2.1
Image EditArena ratingtoday#71,403Muse Image#61,428best of the pickedGemini Nano Banana 2.1
Video2
Text-to-VideoArena ratingtoday#91,456Muse Video#11,516best of the pickedGemini Omni 1.1 Flash
Image-to-VideoArena ratingtoday#21,488Gemini Omni 1.1 Flash
Speech2
Speech ArenaArena ratingtoday#95847Metavoice V1#51,278best of the pickedGemini 3.8 Flash TTS
Open ASRAverage WERtoday#356.48 WERbest of the pickedmuse-voice-transcribe (meta)#5911.65 WERgemma-4-E2B-it (google)
Embeddings1
MTEBMean task scoretoday#568.4%gemini-embedding-001 (google)
Speed1
AA SpeedOutput speedtoday#6345 tok/sBeam 1.0#7281 tok/sMuse Spark 1.3#16209 tok/sDeepSeek V4.1 Flash#4454 tok/sbest of the pickedGemini 3.5 Flash-Lite
Price1
AA PriceBlended pricetoday#15$2.00Muse Spark 1.3#8$0.53best of the pickedDeepSeek V4.1 Flash#13$1.50Gemini 3.8 Flash
Usage1
OpenRouterWeekly tokenstoday#123.3T tokMuse Spark 1.3 Contributor#127.3T 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