{"data":{"slug":"google-deepmind-penzai","name":"penzai","tagline":"A JAX research toolkit for building, editing, and visualizing neural networks.","github_url":"https://github.com/google-deepmind/penzai","owner":"google-deepmind","repo":"penzai","owner_avatar_url":"https://avatars.githubusercontent.com/u/8596759?v=4","primary_language":"Python","stars":1900,"forks":69,"topics":["fine-tuning","interpretability","jax","neural-networks","visualization"],"archived":true,"github_pushed_at":"2025-06-22T05:45:13+00:00","maintenance_label":"Archived","stars_delta_30d":8,"url":"https://www.graphcanon.com/tools/google-deepmind-penzai","markdown_url":"https://www.graphcanon.com/tools/google-deepmind-penzai.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/google-deepmind-penzai","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=google-deepmind-penzai","description":"A JAX research toolkit for building, editing, and visualizing neural networks.","homepage_url":"https://penzai.readthedocs.io/","license":"Apache-2.0","open_issues":21,"watchers":21,"ai_summary":"Penzai is primarily used in the context of model training and visualization within the realm of AI development involving neural networks.","readme_excerpt":"## Getting Started\n\nIf you haven't already installed JAX, you should do that first, since the\ninstallation process depends on your platform. You can find instructions in the\n[JAX documentation](https://jax.readthedocs.io/en/latest/installation.html).\nAfterward, you can install Penzai using\n\n```python\npip install penzai\n```\n\nand import it using\n\n```python\nimport penzai\nfrom penzai import pz\n```\n\n(`penzai.pz` is an *alias namespace*, which makes it easier to reference\ncommon Penzai objects.)\n\nWhen working in an Colab or IPython notebook, we recommend also configuring\nTreescope (Penzai's companion pretty-printer) as the default pretty printer, and\nenabling some utilities for interactive use:\n\n```python\nimport treescope\ntreescope.basic_interactive_setup(autovisualize_arrays=True)\n```\n\nHere's how you could initialize and visualize a simple neural network:\n\n```python\nfrom penzai.models import simple_mlp\nmlp = simple_mlp.MLP.from_config(\n    name=\"mlp\",\n    init_base_rng=jax.random.key(0),\n    feature_sizes=[8, 32, 32, 8]\n)","github_created_at":"2024-04-04T01:13:03+00:00","created_at":"2026-07-11T11:38:40.247208+00:00","updated_at":"2026-09-20T06:29:18.009751+00:00","categories":[{"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":"interpretability","name":"interpretability"},{"slug":"jax","name":"jax"},{"slug":"neural-networks","name":"neural-networks"}],"trust":{"provenance":{"is_fork":false,"github_id":781756197,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-19T18:24:11.142Z","maintenance":{"label":"Archived","score":8,"methodology":"github_public_v1","releases_90d":0,"days_since_push":454,"last_release_at":"2025-04-08T05:45:15Z","stars_delta_30d":8,"open_issues_delta_30d":4},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-11T11:38:41.358Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-09-19T18:24:12.164Z"},"languages":{"value":["python"],"source":"github.language+pyproject.toml","observed_at":"2026-09-19T18:24:12.164Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-09-19T18:24:12.164Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":{"notes":["Ensure compatibility with Python and JAX libraries as they are crucial for leveraging Penzai's toolkit."],"requires_docker":false},"constraints":{"requires_docker":false},"when_to_use":["When your AI development tasks require detailed visualization capabilities for neural networks, as Penzai integrates advanced visual components tailored to these needs.","If your work involves Python-based model training requiring a robust toolkit for both building and editing neural networks efficiently."],"when_not_to_use":["Avoid if you are strictly working with frameworks that do not support or are incompatible with JAX, as Penzai is specifically designed for use within the JAX ecosystem.","Do not choose Penzai if your project requires a focus on backend model deployment rather than research-oriented functionalities like visualization and interpretability."],"source":"enrich:decision_facts","observed_at":"2026-07-16T19:35:34.648Z"},"constraint_facets":{"requires_docker":false},"decision_summary":[{"label":"Requirements","value":"Ensure compatibility with Python and JAX libraries as they are crucial for leveraging Penzai's toolkit."},{"label":"Adopt for","value":"Penzai supports fine-tuning and interpretability features in neural network research through JAX."},{"label":"License detail","value":"Penzai operates under an Apache-2.0 license, offering permissive rights for software use, modification, and distribution."}]}}