{"data":{"slug":"foundationvision-var","name":"VAR","tagline":"Official implementation of Visual Autoregressive Modeling for scalable image generation","github_url":"https://github.com/FoundationVision/VAR","owner":"FoundationVision","repo":"VAR","owner_avatar_url":"https://avatars.githubusercontent.com/u/151817217?v=4","primary_language":"Jupyter Notebook","stars":8731,"forks":571,"topics":["auto-regressive-model","autoregressive-models","diffusion-models","generative-ai","generative-model","gpt","gpt-2","image-generation","large-language-models","neurips","transformers","vision-transformer"],"archived":false,"github_pushed_at":"2025-11-10T21:42:29+00:00","maintenance_label":"Slowing","stars_delta_30d":4,"url":"https://www.graphcanon.com/tools/foundationvision-var","markdown_url":"https://www.graphcanon.com/tools/foundationvision-var.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/foundationvision-var","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=foundationvision-var","description":"[NeurIPS 2024 Best Paper Award][GPT beats diffusion🔥] [scaling laws in visual generation📈] Official impl. of \"Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction\". An *ultra-simple, user-friendly yet state-of-the-art* codebase for autoregressive image generation!","homepage_url":null,"license":"MIT","open_issues":62,"watchers":105,"ai_summary":"A user-friendly codebase for autoregressive image generation using next-scale prediction techniques","readme_excerpt":"## Installation\n\n1. Install `torch>=2.0.0`.\n2. Install other pip packages via `pip3 install -r requirements.txt`.\n3. Prepare the [ImageNet](http://image-net.org/) dataset\n    <details>\n    <summary> assume the ImageNet is in `/path/to/imagenet`. It should be like this:</summary>\n\n    ```\n    /path/to/imagenet/:\n        train/:\n            n01440764: \n                many_images.JPEG ...\n            n01443537:\n                many_images.JPEG ...\n        val/:\n            n01440764:\n                ILSVRC2012_val_00000293.JPEG ...\n            n01443537:\n                ILSVRC2012_val_00000236.JPEG ...\n    ```\n   **NOTE: The arg `--data_path=/path/to/imagenet` should be passed to the training script.**\n    </details>\n\n5. (Optional) install and compile `flash-attn` and `xformers` for faster attention computation. Our code will automatically use them if installed. See [models/basic_var.py#L15-L30](models/basic_var.py#L15-L30).\n\n---\n\n## License\nThis project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.","github_created_at":"2024-04-01T16:53:18+00:00","created_at":"2026-07-07T17:34:13.566684+00:00","updated_at":"2026-09-20T04:33:49.100637+00:00","categories":[{"slug":"computer-vision","name":"Computer Vision","url":"https://www.graphcanon.com/categories/computer-vision","markdown_url":"https://www.graphcanon.com/categories/computer-vision.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/computer-vision"},{"slug":"model-training","name":"Model Training","url":"https://www.graphcanon.com/categories/model-training","markdown_url":"https://www.graphcanon.com/categories/model-training.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/model-training"}],"tags":[{"slug":"auto-regressive-model","name":"auto-regressive-model"},{"slug":"diffusion-models","name":"diffusion-models"},{"slug":"generative-ai","name":"generative-ai"},{"slug":"gpt","name":"gpt"},{"slug":"image-generation","name":"image-generation"},{"slug":"large-language-models","name":"large-language-models"},{"slug":"transformers","name":"transformers"},{"slug":"vision-transformer","name":"vision-transformer"}],"trust":{"provenance":{"is_fork":false,"github_id":780522250,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-18T08:50:44.262Z","maintenance":{"label":"Slowing","score":36,"methodology":"github_public_v1","releases_90d":0,"days_since_push":311,"last_release_at":null,"stars_delta_30d":4,"open_issues_delta_30d":2},"security_summary":{"status":"ok","scanner":"osv@v1","low_count":0,"high_count":0,"last_scan_at":"2026-09-18T08:50:45.899Z","medium_count":0,"scan_profile":"deps","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-09-18T08:50:43.193Z"},"languages":{"value":["jupyter notebook"],"source":"github.language","observed_at":"2026-09-18T08:50:43.193Z"},"license_spdx":{"value":"MIT","source":"github.license","observed_at":"2026-09-18T08:50:43.193Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":{"notes":["Requires installation of `torch>=2.0.0` and other dependencies listed in `requirements.txt`.","Optional installation of `flash-attn` and `xformers` for faster attention computation."],"min_ram_gb":8,"requires_docker":false},"constraints":{"min_ram_gb":8,"requires_docker":false},"when_to_use":["When you need a state-of-the-art method for autoregressive image generation that is both simple and user-friendly.","If you are working with large datasets like ImageNet and require scalable image generation capabilities.","When you want to leverage next-scale prediction techniques for generating high-quality images.","If you are interested in using the latest advancements in visual autoregressive modeling, as recognized by the NeurIPS 2024 Best Paper Award."],"when_not_to_use":["If you require real-time image generation capabilities, as VAR focuses on scalable and high-quality image generation rather than speed.","When your project does not involve image generation tasks, as VAR is specifically designed for autoregressive image generation.","If you are looking for a tool that supports a wide range of generative models beyond autoregressive modeling, as VAR is specialized in this area.","When you need a tool that does not require the ImageNet dataset for training, as VAR assumes the use of this dataset for its operations."],"source":"enrich:decision_facts","observed_at":"2026-09-18T10:54:12.945Z"},"constraint_facets":{"min_ram_gb":8,"requires_docker":false},"decision_summary":[{"label":"Requirements","value":"Min 8 GB RAM; Requires installation of `torch>=2.0.0` and other dependencies listed in `requirements.txt`.; Optional installation of `flash-attn` and `xformers` for faster attention computation."},{"label":"Adopt for","value":"VAR is an autoregressive image generation tool that uses next-scale prediction techniques, offering a user-friendly codebase for scalable image generation."}]}}