Skip to main content

MuJoCo · Core16

Reacher

Continuous robotic-arm control toward a target point.

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 Reacherdirect mujoco renderer

GPT-5.5

Codex
BEST
Held-out
-3.473
Checkpoint
#012
Rerun
50 steps

Validation-selected checkpoint rerun in the original environment; case return -1.880.

Claude Opus 4.7 Policy rerun in Reacherdirect mujoco renderer

Claude Opus 4.7

Claude Code
#2
Held-out
-3.979
Checkpoint
#007
Rerun
50 steps

Validation-selected checkpoint rerun in the original environment; case return -2.670.

MiniMax-M3 Policy rerun in Reacherdirect mujoco renderer

MiniMax-M3

Claude Code
#3
Held-out
-5.103
Checkpoint
#015
Rerun
50 steps

Validation-selected checkpoint rerun in the original environment; case return -2.308.

DeepSeek-V4-Pro Policy rerun in Reacherdirect mujoco renderer

DeepSeek-V4-Pro

Claude Code
#4
Held-out
-6.506
Checkpoint
#019
Rerun
50 steps

Validation-selected checkpoint rerun in the original environment; case return -4.613.

02

Reported scores

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

01GPT-5.5Codex-3.473
02Claude Opus 4.7Claude Code-3.979
03MiniMax-M3Claude Code-5.103
04DeepSeek-V4-ProClaude Code-6.506
05Random policyUniform-43.77