PRIME-RL/SimpleVLA-RL
Open RL framework for training Vision-Language-Action models to improve long-horizon planning and task completion.
Not currently ranked — collecting fresh signals.
star history
SimpleVLA-RL is a reinforcement learning framework designed to scale Vision-Language-Action model training, particularly under data scarcity conditions. It applies RL-based fine-tuning to VLAs that can perceive visual inputs, process language instructions, and generate actions for robotic or embodied AI tasks. The framework reportedly substantially outperforms supervised fine-tuning on both simulation and real-world tasks.
Frequently asked
- What is PRIME-RL/SimpleVLA-RL?
- Open RL framework for training Vision-Language-Action models to improve long-horizon planning and task completion.
- Is SimpleVLA-RL open source?
- Yes — PRIME-RL/SimpleVLA-RL is open source, released under the MIT license.
- What language is SimpleVLA-RL written in?
- PRIME-RL/SimpleVLA-RL is primarily written in Python.
- How popular is SimpleVLA-RL?
- PRIME-RL/SimpleVLA-RL has 1.8k stars on GitHub.
- Where can I find SimpleVLA-RL?
- PRIME-RL/SimpleVLA-RL is on GitHub at https://github.com/PRIME-RL/SimpleVLA-RL.