Read, search, annotate, and ask questions about PDF, EPUB, and DOCX textbooks in one private Windows app.
Download for Windows · English · 简体中文 · 繁體中文
- One focused library: import PDF, EPUB, and DOCX books without juggling separate readers.
- Persistent study context: your books, reading position, notes, annotations, and completed question history stay organized by textbook.
- Less chat fragmentation: ordinary AI chats have finite context, so long study sessions often spread across new conversations and repeated uploads. TextbookLens retrieves relevant current-book material for each question. It does not remove a model's context limit, but it reduces repeated setup and keeps the learning trail together.
- Local-first by design: reading, local search, annotations, history, and backups remain on your device.
- AI only when you ask: optional AI operations start only from an explicit action and send the bounded content needed for that operation.
- Offline local AI: connect installed Ollama text models with one action. Compatible Ollama vision models can answer questions about a selected image region. LM Studio text support remains experimental.
- Open and modifiable: the Rust, React, Tauri, and SQLite source is available under Apache-2.0.
Actual application interface shown with synthetic test data.
Open the latest release and choose:
TextbookLens_*_x64-setup.exe— recommended small installer; uses the existing WebView2 runtime or downloads it when missing.TextbookLens_*_x64_en-US.msi— standard MSI for managed installation.- A file containing
offline— complete offline installer with the Microsoft WebView2 runtime included. SHA256SUMS.txt— checksums for all four installers.
Neither package type includes Ollama, LM Studio, or model weights. After installation, import a PDF, EPUB, or DOCX. AI is optional: connect an installed local runtime or configure a supported cloud provider only when needed.
Important
Current packages support Windows 11 x64 and are not code-signed. Windows SmartScreen may show an unknown-publisher warning. Verify the SHA-256 before choosing Run anyway.
TextbookLens has no account system, cloud sync, bundled AI credit, project proxy, collaboration service, telemetry, or silent background upload. Cloud API keys are stored through Windows Credential Manager rather than the application database, browser storage, logs, diagnostics, or backups. Local Ollama and LM Studio profiles do not require a cloud key.
Importing, reading, local search, annotations, language changes, backup, and restore do not require an AI request. See AI services and sending boundary, Offline local models, and Data, backup, restore, and deletion for exact behavior.
git clone https://github.com/gh615280-maker/TextbookLens.git
cd TextbookLens
npm ci
npm run tauri devThe pinned toolchain is Node.js 24.18.1, npm 11.16.0, Rust 1.97.1, and the Windows prerequisites for Tauri 2. Run npm run preflight for release-oriented validation and npm run tauri build for local NSIS/MSI packages.
Read CONTRIBUTING.md before changing data, credentials, networking, migrations, fixtures, or generated bindings. Dependency provenance is recorded in LICENSES.md and THIRD_PARTY_NOTICES.md.
Version 0.2.0 is the current early public release. It adds one-action Ollama discovery, offline text Q&A, independent Ollama vision selection, bundled PDF decoding resources, and reader reliability fixes. The release passed the repository's frontend, Rust, security, Windows build, installer, offline-install, and v0.1.0 upgrade checks documented in the v0.2.0 validation record.
Known limits remain: packages are unsigned, automatic updates are not implemented, LM Studio real-machine text validation is deferred to v0.2.1, local whole-book OCR/vision indexing is not implemented, and local vision answers can be factually wrong on difficult material. See the v0.2.0 release notes.
Ideas and bug reports are welcome through GitHub Issues. If the project is useful, star the repository so more readers and contributors can find it.