Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.14993 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914603789713408 |
|---|---|
| author | Arkhipkin, Vladimir Korviakov, Vladimir Gerasimenko, Nikolai Parkhomenko, Denis Vasilev, Viacheslav Letunovskiy, Alexey Vaulin, Nikolai Kovaleva, Maria Kirillov, Ivan Novitskiy, Lev Koposov, Denis Kiselev, Nikita Varlamov, Alexander Mikhailov, Dmitrii Polovnikov, Vladimir Shutkin, Andrey Agafonova, Julia Vasiliev, Ilya Kargapoltseva, Anastasiia Dmitrienko, Anna Maltseva, Anastasia Averchenkova, Anna Kim, Olga Nikulina, Tatiana Dimitrov, Denis |
| author_facet | Arkhipkin, Vladimir Korviakov, Vladimir Gerasimenko, Nikolai Parkhomenko, Denis Vasilev, Viacheslav Letunovskiy, Alexey Vaulin, Nikolai Kovaleva, Maria Kirillov, Ivan Novitskiy, Lev Koposov, Denis Kiselev, Nikita Varlamov, Alexander Mikhailov, Dmitrii Polovnikov, Vladimir Shutkin, Andrey Agafonova, Julia Vasiliev, Ilya Kargapoltseva, Anastasiia Dmitrienko, Anna Maltseva, Anastasia Averchenkova, Anna Kim, Olga Nikulina, Tatiana Dimitrov, Denis |
| contents | This report introduces Kandinsky 5.0, a family of state-of-the-art foundation models for high-resolution image and 10-second video synthesis. The framework comprises three core line-up of models: Kandinsky 5.0 Image Lite - a line-up of 6B parameter image generation models, Kandinsky 5.0 Video Lite - a fast and lightweight 2B parameter text-to-video and image-to-video models, and Kandinsky 5.0 Video Pro - 19B parameter models that achieves superior video generation quality. We provide a comprehensive review of the data curation lifecycle - including collection, processing, filtering and clustering - for the multi-stage training pipeline that involves extensive pre-training and incorporates quality-enhancement techniques such as self-supervised fine-tuning (SFT) and reinforcement learning (RL)-based post-training. We also present novel architectural, training, and inference optimizations that enable Kandinsky 5.0 to achieve high generation speeds and state-of-the-art performance across various tasks, as demonstrated by human evaluation. As a large-scale, publicly available generative framework, Kandinsky 5.0 leverages the full potential of its pre-training and subsequent stages to be adapted for a wide range of generative applications. We hope that this report, together with the release of our open-source code and training checkpoints, will substantially advance the development and accessibility of high-quality generative models for the research community. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_14993 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Kandinsky 5.0: A Family of Foundation Models for Image and Video Generation Arkhipkin, Vladimir Korviakov, Vladimir Gerasimenko, Nikolai Parkhomenko, Denis Vasilev, Viacheslav Letunovskiy, Alexey Vaulin, Nikolai Kovaleva, Maria Kirillov, Ivan Novitskiy, Lev Koposov, Denis Kiselev, Nikita Varlamov, Alexander Mikhailov, Dmitrii Polovnikov, Vladimir Shutkin, Andrey Agafonova, Julia Vasiliev, Ilya Kargapoltseva, Anastasiia Dmitrienko, Anna Maltseva, Anastasia Averchenkova, Anna Kim, Olga Nikulina, Tatiana Dimitrov, Denis Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning This report introduces Kandinsky 5.0, a family of state-of-the-art foundation models for high-resolution image and 10-second video synthesis. The framework comprises three core line-up of models: Kandinsky 5.0 Image Lite - a line-up of 6B parameter image generation models, Kandinsky 5.0 Video Lite - a fast and lightweight 2B parameter text-to-video and image-to-video models, and Kandinsky 5.0 Video Pro - 19B parameter models that achieves superior video generation quality. We provide a comprehensive review of the data curation lifecycle - including collection, processing, filtering and clustering - for the multi-stage training pipeline that involves extensive pre-training and incorporates quality-enhancement techniques such as self-supervised fine-tuning (SFT) and reinforcement learning (RL)-based post-training. We also present novel architectural, training, and inference optimizations that enable Kandinsky 5.0 to achieve high generation speeds and state-of-the-art performance across various tasks, as demonstrated by human evaluation. As a large-scale, publicly available generative framework, Kandinsky 5.0 leverages the full potential of its pre-training and subsequent stages to be adapted for a wide range of generative applications. We hope that this report, together with the release of our open-source code and training checkpoints, will substantially advance the development and accessibility of high-quality generative models for the research community. |
| title | Kandinsky 5.0: A Family of Foundation Models for Image and Video Generation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2511.14993 |