Library
Browse by model and by use case
The arena compares models on one challenge. The library lets you inspect a single model across every challenge, or a challenge across every model.
By model
xAI
Grok 4.6
7 published challenges
OpenAIGPT-5.6 Luna
7 published challenges
GoogleGemini 3.8 Flash
7 published challenges
Z.aiGLM 5.3 Flash
7 published challenges
TencentHY 4 Preview
7 published challenges
DeepSeekDeepSeek V4 Flash 0731
7 published challenges
Moonshot AIKimi K3
7 published challenges
OpenAIGPT-5.6 Sol
7 published challenges
MetaMuse Spark 1.3
7 published challenges
GoogleGemma 4 31B IT
7 published challenges
By use case
Geometry / depth
Solar System
10 model builds
Complexity / agentsCity Traffic
10 model builds
Organic motionLiving Aquarium
10 model builds
Visual causalityRube Goldberg
10 model builds
System coordinationFuture Factory
10 model builds
Autonomous platformingSuper Mario Bros
10 model builds
Digital rain / densityMatrix Screen
10 model builds