The field of generative modeling is witnessing a significant shift towards the integration of flow-based architectures and biological applications. Recent developments have led to the creation of novel generative flow models, such as Gauge Flow Models and Higher Gauge Flow Models, which have shown substantial performance improvements over traditional flow models. These models incorporate learnable gauge fields and higher geometry, allowing for the integration of complex symmetries and structures into the generative process. In the context of biology, generative models are being increasingly used to explore protein conformational spaces, design novel molecules, and generate functional protein sequences. Techniques such as flow matching and conditional diffusion models are being applied to biological problems, enabling breakthroughs in molecule design, protein generation, and drug discovery. Noteworthy papers in this area include Gauge Flow Models, which introduced a novel class of generative flow models with improved performance, and MoDyGAN, which combined molecular dynamics with GANs to explore protein conformational spaces. Higher Gauge Flow Models also demonstrated substantial performance improvements over traditional flow models, while Demystify Protein Generation with Hierarchical Conditional Diffusion Models proposed a novel multi-level conditional diffusion model for efficient end-to-end protein design. These developments highlight the potential of generative models to drive innovation in biology and life sciences.