Every AI leaderboard

Leaderboards28Models ranked986

28 leaderboards · 986 models ranked

StepFun 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.

  • StepFun
  • OpenAI
Add:AnthropicGoogleMicrosoftDeepSeek

Head to head on 8 boards: OpenAI places best on 7 and StepFun places best on 1. Both ranked on 8 of 26 boards.

BoardStepFunOpenAI
Chat2
Text ArenaArena ratingtoday#701,448Step 5 Preview#191,485best of the pickedGPT-5.6 Sol
MultiChallengeScoretoday#182.0%GPT-6.1 Sol
Reasoning7
AA IntelligenceIndextoday#2328.5Step 5 Preview#356.9best of the pickedGPT-6 Astra
HLEAccuracytoday#154.8%GPT-6 Astra
Epoch ECIECItoday#1169.1GPT-6 Astra
ARC-AGI-2Scoretoday#195.0%GPT-6 Astra
GPQA DiamondAccuracytoday#295.8%GPT-6 Astra
MMLU-ProAccuracy · staletoday#389.3%o1
LiveBenchGlobal averagetoday#482.2%GPT-6 Astra
Coding4
WebDev ArenaArena ratingtoday#381,572Step 5 Preview#31,770best of the pickedGPT-6 Astra
Aider PolyglotPass rate · staletoday#188.0%GPT-5
SWE-Bench ProResolvedtoday#490.2%GPT-6 Astra
SWE-bench VerifiedResolved · staletoday#672.8%GPT-5.2
Agents and tools3
Search ArenaArena ratingtoday#11,255GPT-5.6 Sol
BFCLOverall accuracy · staletoday#763.0%o3
MCP AtlasPass ratetoday#1481.8%GPT-5.6
Vision2
Vision ArenaArena ratingtoday#511,247Step 5 Preview#81,293best of the pickedGPT-6.1 Sol
Document ArenaArena ratingtoday#51,486GPT-5.6 Sol
Image2
Image EditArena ratingtoday#51996Step1x Edit#11,517best of the pickedGPT Image 2.5 Sunburst
Text-to-ImageArena ratingtoday#11,429GPT Image 2.5 Sunburst
Video1
Text-to-VideoArena ratingtoday#131,367Sora 2 Pro
Speech2
Speech ArenaArena ratingtoday#181,216best of the pickedStepaudio 2.5 TTS#421,107TTS 1 Hd
Open ASRAverage WERtoday#186.80 WERgpt-transcribe (openai)
Speed1
AA SpeedOutput speedtoday#2214 tok/sStep 5 Preview#1522 tok/sbest of the pickedGPT-6.1 Sol
Price1
AA PriceBlended pricetoday#6$2.00GPT-6.1 Terra
Usage1
OpenRouterWeekly tokenstoday#57.2T tokStep 5 Preview#47.2T tokbest of the pickedGPT-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