open-science
AIPOCH Open-Science is an open-source, local-first research workbench (Apache 2.0) engineered for scientists, computational researchers, and data practitioners across machine learning, life sciences, physics, and materials engineering. Designed around local data privacy and reproducible methodology, it provides a transparent alternative to opaque cloud AI research tools.
The workbench orchestrates autonomous scientific agents across multiple frameworks (supporting Pi and OpenCode), backed by isolated Python and R execution environments. Researchers can ingest academic papers directly via DOI or arXiv links, read and annotate split-screen PDFs with multi-lingual OCR, execute data analysis pipelines, and generate publication-ready tables and figures with traceable computational provenance.
A standout feature is the exportable '.science' bundle format, which encapsulates conversation threads, source code versions, Jupyter notebook sessions, and verification artifacts into a portable archive for peer review and collaboration. While exceptionally capable, the local desktop application demands significant local memory and compute when running local OCR or executing multi-step computational simulations.
Features
Why choose it
Trade-offs
Maintenance check
- Last commit
- 2026-11-12
- Commits, last 90 days
- 100+
- Releases, last 90 days
- 30
- Latest release
- v1.2.1 (2026-11-11)
- Open issues
- 36
- Recent issues with at least one comment
- 0/7
Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.
Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.
Most discussed open issues
- #3302 [Feedback]: Allow assigning new projects to custom folders for file reading and output generation1 comments · 2026-10-05
- #925 [Feature]: RO-Crate export of artifact provenance1 comments · 2026-08-07
- #4308 ci(windows-full-test.yml): scheduled run failing0 comments · 2026-11-12
- #4302 ci(source-regression.yml): scheduled run failing0 comments · 2026-11-11
- #2728 [Reproducibility Pilot] Review the first 10 cases and verify an improvement0 comments · 2026-09-17
Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.
Compatibility
Quick start
Use cases
Alternatives
Comments
No comments yet. Be the first!