GPT4All
Nomic AICross-platform desktop app from Nomic AI for running LLMs locally on ordinary laptops, without an API key or a GPU. It bundles a chat interface, LocalDocs for retrieval over private files, an OpenAI-compatible local API server and a Python client, with Nomic Vulkan acceleration where a GPU is available.
Development has stalled: the repository has had no push since 27 May 2025 and there is no official deprecation notice.
Openness
5 high confidence- license
- MIT(OSI)
- source
- public
- core features withheld
- no
Fully MIT. The repository ships the whole application - installers for Windows, Windows ARM, macOS and Ubuntu plus the Python client - with nothing withheld and no paid tier.
- https://github.com/nomic-ai/gpt4all recorded 2026-08-13
MIT, ~77.4k stars, LocalDocs RAG + local API server
- https://nomic.ai/gpt4all recorded 2026-09-29
The GPT4All page answers "Is GPT4All free to use?" with "Yes, GPT4All is completely free and open-source. You can download and run it on your computer at no cost, with no subscription fees or API charges."
- https://pypi.org/pypi/gpt4all/json recorded 2026-09-29
The PyPI record for the gpt4all Python bindings lists the classifier "License :: OSI Approved :: MIT License" and a source-code link to "https://github.com/nomic-ai/gpt4all/tree/main/gpt4all-bindings/python".
Adoption
2 medium confidenceAdoption is measured by combined downloads of the gpt4all PyPI package and its Node bindings, the two published distribution channels. Because a real download signal exists, it is used instead of the GitHub star count, which reads higher than actual use here.
- https://api.npmjs.org/downloads/point/last-month/gpt4all recorded 2026-08-12
1,528 downloads in the trailing 30 days for gpt4all
- https://pypistats.org/api/packages/gpt4all/recent recorded 2026-08-12
58,808 downloads in the trailing 30 days for gpt4all
Capability
3 medium confidenceSolid core but behind faster-moving peers on newest-model support due to the maintenance slowdown.
- https://docs.gpt4all.io/ recorded 2026-09-29
The documentation home describes a desktop application with "No API calls or GPUs required - you can just download the application and " a Python SDK installed with pip.
- https://docs.gpt4all.io/gpt4all_api_server/home.html recorded 2026-09-29
The API Server page states "GPT4All provides a local API server that allows you to run LLMs over an HTTP API." and lists OpenAI API Compatibility and LocalDocs Integration as key features.
- https://docs.gpt4all.io/gpt4all_desktop/localdocs.html recorded 2026-09-29
The LocalDocs page says the feature "brings the information you have from files on-device into your LLM chats - " privately, indexing a folder with on-device embedding models.
- https://docs.gpt4all.io/gpt4all_help/faq.html recorded 2026-09-29
The FAQ states "GPT4All can run on CPU, Metal (Apple Silicon M1+), and GPU."
- https://github.com/nomic-ai/gpt4all recorded 2026-08-13
feature set: chat, LocalDocs, local server
Verified 2026-09-29, except some scores, still awaiting re-confirmation