OpenGameEval: Eval Framework to Benchmark Agentic AI Assistants(corp.roblox.com)
corp.roblox.com
OpenGameEval: Eval Framework to Benchmark Agentic AI Assistants
https://corp.roblox.com/newsroom/2025/12/opengameeval-benchmark-agentic-ai-assistants-roblox-studio
1 comments
OpenGameEval offers a unique testing ground to evaluate core model capabilities related to agenetic reasoning and long-horizon task solving.