NeuralScio has a meaningful local path, but offline work centers on saved-library continuity: cached PDFs, saved metadata, local annotations, and local AI where configured.
What Works Offline
| Feature | Offline support | | --- | :---: | | Reading cached PDFs | Available | | Browsing saved articles | Available | | Viewing cached web pages | Available | | Local AI when configured | Available | | Citation graph on already available data | Available | | Collections and local library work | Available | | Searching external providers | Not available | | Managed cloud AI | Not available | | Account-linked sync | Experimental; conditional on the current build and account configuration | | Downloading new remote content | Not available |
What The Boundary Means
The desktop application includes:
- a local PDF library
- offline prefetch and cache services for saved content
- locally stored annotations and metadata
- the built-in NeuralScioAI worker with GGUF models, plus the separate Ollama-compatible provider path
Saved content remains usable offline, but provider search and cloud services still require the network. Sync is experimental and conditional on the current build and account configuration.
Preparing For Offline Work
- Search and save the papers you need while online.
- Open or prefetch the papers you expect to need locally.
- Keep local models configured if you want AI help without the network.
- Treat sync as an optional, experimental companion layer, not as the only copy of your work.
Practical Limits
- External provider search still needs the network.
- Managed cloud AI still needs the network.
- If sync is enabled for the current build and account, it requires connectivity; do not assume that every account has this service available.
- Offline continuity is strongest for already saved and cached material.
Related Docs
- Read PDF Viewer for the reading surface.
- Read AI Features for local-vs-cloud model choices.
- Read MCP & Headless if your local workflow also includes agents or scripts.