Skip to main content

MiniGrid · Core16

ObstructedMaze

Maze navigation with blockers and interaction sequences.

Historical research record

This evidence belongs to the paper-era experiment and is not a guarantee of current package availability.

01

Final-Policy evidence

Each card reruns the validation-selected checkpoint in the original research Environment.

GPT-5.5 Policy rerun in ObstructedMazenative env render

GPT-5.5

Codex
#2
Held-out
0.904
Checkpoint
#021
Rerun
90 steps

Validation-selected checkpoint rerun in the original environment; case return 0.887.

Claude Opus 4.7 Policy rerun in ObstructedMazenative env render

Claude Opus 4.7

Claude Code
BEST
Held-out
0.911
Checkpoint
#012
Rerun
48 steps

Validation-selected checkpoint rerun in the original environment; case return 0.940.

MiniMax-M3 Policy rerun in ObstructedMazenative env render

MiniMax-M3

Claude Code
#3
Held-out
0.000
Checkpoint
#009
Rerun
220 steps

Validation-selected checkpoint rerun in the original environment; case return 0.000.

DeepSeek-V4-Pro Policy rerun in ObstructedMazenative env render

DeepSeek-V4-Pro

Claude Code
#4
Held-out
0.000
Checkpoint
#034
Rerun
220 steps

Validation-selected checkpoint rerun in the original environment; case return 0.000.

02

Reported scores

Higher is better within this Environment. Raw reward scales are not comparable across tasks.

01Claude Opus 4.7Claude Code0.911
02GPT-5.5Codex0.904
03MiniMax-M3Claude Code0.000
04DeepSeek-V4-ProClaude Code0.000
05Random policyUniform0.000