TerrariaBench / run

GPT-5.5

api controlling Open-ended progress for 28802s. Best visible milestone: held pickaxe.

Open replay Result JSON All visual runs Raw data
Harness
api
Model
GPT-5.5
Elapsed
28802s
Agency Score
0

Side-by-side replay preview

Open MP4

Run Summary

Status
invalid
Milestone
held pickaxe
Frames
3080
Raw Checkpoints
3
Survival Checkpoints
0
Furthest Level
L0
Actions
4932

run_timelapse.mp4

143 KB