openclaw-rl-training
reason-machines1.3k installs
Train your AI agent from everyday conversations.
What it does
Trains personalized AI agents using reinforcement learning signals derived from live conversations.
When to use it
Use when setting up agent training from conversation feedback or configuring an asynchronous reinforcement learning pipeline.
How to use it
Give it a model and conversation data; it configures and launches agent training with a checkpoint destination.
What you provide
- Your files
Uses
TI
Latest review
What you need · 4
The API server is launched with Python; no Python version is specified.
The local training configuration partitions GPUs between training and rollout; the eight-GPU layout is explicitly an example.
Access · 3
This skill
slime
Execute
Installs the slime training backend
OpenClaw-RL
Execute
Starts continuous background agent training