Roblox
We track 3 benchmarks that Roblox made or helped make.
0 in use 3 with public code 3 with a leaderboard here
Benchmarks
| Name | Kind | Released | Best score | Status |
|---|---|---|---|---|
| OpenGameEval Agents write Luau scripts and edit objects in Roblox Studio. The original suite checks 47 changes to game mechanics and scenes. | Building whole projects | Oct 2025 | | Not in cards |
| OpenGameEval Debug Evals Agents diagnose and fix faults in Roblox Studio games. The authors created 30 debugging tasks from 15 scenarios. | Fixing real repositories | Mar 2026 | 64.7% | Not in cards |
| OpenGameEval Expanded Eval Set Agents extend existing Roblox games and build features from empty scenes. Roblox added 40 coding tasks to the original 47. | Building whole projects | Apr 2026 | 50.3% | Not in cards |