Flash-RL by yaof20

Multimodal RL training framework for diffusion & omni models. Maintained by verl-project on GitHub, where it has earned 728 stars from the community.

UniRL is a Framework for Unified Multimodal Model Reinforcement Learning. Maintained by Tencent-Hunyuan on GitHub, where it has earned 881 stars from the community.

Synthetic medical VQA pipeline: 119K images annotated by frontier VLMs, cross-validated at 93% agreement, fine-tuned on 3 model families (2-3B params). Maintained by openmed-labs on GitHub, where it has earned 39 stars from the community.