Tencent-Hunyuan-Multimodal-RL/SD3.5-GenEval2-Single-Reward
Text-to-Image • Updated • 4
Flow-DPPO: GenEval2
Flow-DPPO-trained LoRA adapters (single- and multi-reward) for SD3.5 and FLUX.2-klein-9B optimized on GenEval2.
-
-
Tencent-Hunyuan-Multimodal-RL/SD3.5-GenEval2-Multi-Reward
Text-to-Image • Updated • 3 -
Tencent-Hunyuan-Multimodal-RL/FLUX2-klein-base-9b-GenEval2-Single-Reward
Text-to-Image • Updated • 11 • 1 -
Tencent-Hunyuan-Multimodal-RL/FLUX2-klein-base-9b-GenEval2-Multi-Reward
Text-to-Image • Updated • 21 • 3 -
Flow-DPPO: Divergence Proximal Policy Optimization for Flow Matching Models
Paper • 2606.11025 • Published • 41