Genmo is an artificial intelligence research company headquartered in San Francisco, focused on building open-source generative video foundation models. The company gained prominence with the October 2024 release of Mochi 1, a 10-billion-parameter text-to-video diffusion model that it describes as the largest open video model released to that point, with weights and architecture available on Hugging Face.
Mochi 1 is built on Genmo's proprietary Asymmetric Diffusion Transformer (AsymmDiT) architecture. The company claims the model delivers motion quality and prompt adherence competitive with leading closed systems such as Runway Gen-3, Luma Dream Machine, Kling and Minimax Hailuo, while remaining open for researchers and builders to fine-tune and deploy.
Genmo was founded in 2022 by brothers Paras Jain and Ajay Jain, both machine-learning researchers with academic backgrounds in efficient deep-learning systems and diffusion models. Paras Jain serves as CEO and Ajay Jain as CTO.
In October 2024 the company announced a $28.4 million Series A round led by NEA, with participation from The House Fund, Gold House Ventures, WndrCo, Eastlink Capital Partners and Essence VC, plus angel investors including Replit CEO Amjad Masad. The capital is directed at scaling video model training and building products on top of its open models.
Genmo's strategy contrasts with the predominantly closed approach of major video AI labs, betting that open weights will drive faster ecosystem adoption among developers, studios and academic researchers. The company continues to iterate on the Mochi line and related image and 3D generation capabilities.