{"data":{"slug":"physical-intelligence-openpi","name":"openpi","tagline":"Repository for running AI models with GPU requirements specified for inference and fine-tuning.","github_url":"https://github.com/Physical-Intelligence/openpi","owner":"Physical-Intelligence","repo":"openpi","owner_avatar_url":"https://avatars.githubusercontent.com/u/162759805?v=4","primary_language":"Python","stars":13874,"forks":2467,"topics":[],"archived":false,"github_pushed_at":"2026-08-24T16:27:13+00:00","maintenance_label":"Active","stars_delta_30d":776,"url":"https://www.graphcanon.com/tools/physical-intelligence-openpi","markdown_url":"https://www.graphcanon.com/tools/physical-intelligence-openpi.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/physical-intelligence-openpi","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=physical-intelligence-openpi","description":null,"homepage_url":null,"license":"Apache-2.0","open_issues":345,"watchers":99,"ai_summary":"This repository contains code for running AI models, with a focus on specifying GPU requirements for different modes of operation, including inference and fine-tuning with LoRA and full fine-tuning.","readme_excerpt":"## Requirements\n\nTo run the models in this repository, you will need an NVIDIA GPU with at least the following specifications. These estimations assume a single GPU, but you can also use multiple GPUs with model parallelism to reduce per-GPU memory requirements by configuring `fsdp_devices` in the training config. Please also note that the current training script does not yet support multi-node training.\n\n| Mode               | Memory Required | Example GPU        |\n| ------------------ | --------------- | ------------------ |\n| Inference          | > 8 GB          | RTX 4090           |\n| Fine-Tuning (LoRA) | > 22.5 GB       | RTX 4090           |\n| Fine-Tuning (Full) | > 70 GB         | A100 (80GB) / H100 |\n\nThe repo has been tested with Ubuntu 22.04, we do not currently support other operating systems.\n\n---\n\n## Installation\n\nWhen cloning this repo, make sure to update submodules:\n\n```bash\ngit clone --recurse-submodules git@github.com:Physical-Intelligence/openpi.git","github_created_at":"2024-10-21T15:23:28+00:00","created_at":"2026-07-11T23:10:04.061015+00:00","updated_at":"2026-09-20T04:29:30.636755+00:00","categories":[{"slug":"inference-serving","name":"Inference & Serving","url":"https://www.graphcanon.com/categories/inference-serving","markdown_url":"https://www.graphcanon.com/categories/inference-serving.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/inference-serving"},{"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":"fine-tuning","name":"fine-tuning"},{"slug":"gpu","name":"gpu"},{"slug":"inference","name":"inference"},{"slug":"lora","name":"lora"},{"slug":"model-parallelism","name":"model parallelism"},{"slug":"nvidia","name":"nvidia"},{"slug":"python","name":"python"},{"slug":"ubuntu-22-04","name":"ubuntu 22.04"}],"trust":{"provenance":{"is_fork":false,"github_id":876210014,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-18T08:16:17.376Z","maintenance":{"label":"Active","score":82,"methodology":"github_public_v1","releases_90d":0,"days_since_push":24,"last_release_at":null,"stars_delta_30d":776,"open_issues_delta_30d":29},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-09-18T08:16:18.988Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-09-18T08:16:16.267Z"},"languages":{"value":["python"],"source":"github.language+pyproject.toml","observed_at":"2026-09-18T08:16:16.267Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-09-18T08:16:16.267Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":{"notes":["The repository requires an NVIDIA GPU with at least 8 GB of memory for inference, 22.5 GB for fine-tuning with LoRA, and 70 GB for full fine-tuning."],"min_ram_gb":null,"requires_docker":false},"constraints":{"min_ram_gb":null,"requires_docker":false},"when_to_use":["Use openpi when you need to run AI models with precise GPU memory requirements for inference and fine-tuning, particularly with LoRA and full fine-tuning modes.","Consider openpi if you are working with Ubuntu 22.04 and have access to GPUs like the RTX 4090 or A100 (80GB) / H100, as it has been tested with these configurations."],"when_not_to_use":["Avoid using openpi if your project requires multi-node training, as the current training script does not support this feature.","Do not use openpi if you are working on an operating system other than Ubuntu 22.04, as the repository has not been tested with other systems."],"source":"enrich:decision_facts","observed_at":"2026-09-18T09:46:01.699Z"},"constraint_facets":{"min_ram_gb":null,"requires_docker":false},"decision_summary":[{"label":"Requirements","value":"The repository requires an NVIDIA GPU with at least 8 GB of memory for inference, 22.5 GB for fine-tuning with LoRA, and 70 GB for full fine-tuning."},{"label":"Adopt for","value":"openpi is a repository for running AI models with specific GPU requirements for inference and fine-tuning, including LoRA and full fine-tuning modes."}]}}