🎬RL GIFs

Trained agents in action—watch SLM Lab's PPO and SAC algorithms play games and control robots.

These GIFs show trained agents running in "enjoy" mode. Generate your own with:

slm-lab run slm_lab/spec/benchmark/ppo/ppo_cartpole.json ppo_cartpole enjoy@data/ppo_cartpole_2026_01_30_221924/ppo_cartpole_t0_spec.json

Atari (PPO)

BeamRider	Breakout
KungFuMaster	MsPacman
Pong	Qbert
Seaquest	SpaceInvaders

MuJoCo (SAC)

Ant	HalfCheetah
Hopper	Humanoid
InvertedDoublePendulum	InvertedPendulum
Reacher	Walker2d

These replays were generated with SLM Lab v4 using Roboschool. In v5, MuJoCo environments (Hopper-v5, HalfCheetah-v5, etc.) provide similar physics with improved stability.