textual Releases
166 releases of Textualize/textual
- v6.4.0The one word command release
Focuses on dependency management fixes and simplifying the user interface of the command palette.
Oct 22, 2025
- v6.3.0The Pithonic release
Updates Python support and adds a new CSS rule for scrollbar visibility.
Oct 11, 2025
- v6.2.1The Copy release
A hotfix release addressing two specific issues related to text copying functionality.
Oct 1, 2025
- v6.2.0The Eager release
A mixed bag of features and fixes, including enabling eager tasks on Python 3.12+, making layout methods public, and improving hover pseudo-class behavior.
Sep 30, 2025
- v0.17.0New Layout Model
This release introduces a completely new architecture for the layout model, trained from scratch, along with a new unified tokenizer to improve performance across various domains.
Sep 23, 2025
- 25.9.0Release 25.9.0
This release adds extensive new hardware support for RTX Pro 6000 and DGX B200, introduces new VLM models for OCR and captioning, and enables custom vector database implementations and metadata types.
Sep 17, 2025
- v0.16.7Move flash imports
This release focuses on internal refactoring by moving Flash Attention functions to improve code organization and import structure.
Sep 8, 2025
- v0.16.6Enable setting attention method
The release enables users to configure the attention method directly and removes strict checks, offering more flexibility in how attention is calculated.
Sep 8, 2025
- v6.1.0The Flat release
Introduces visual styling updates including a new 'block' border style and flat button styles, along with a new parameter for action running.
Sep 2, 2025
- v3.11.2
This release fixes critical input bugs for subscripts and superscripts, improves security by addressing XSS in markdown pasting, and adds new features like custom toolbar configuration and support for Spanish/Vietnamese languages.
Sep 2, 2025
- v6.0.0The Anniversary Release
A large update featuring optimizations to the line API, a new `Binding.Group` for key management, and various improvements to TextArea and Header widgets.
Aug 31, 2025
- v6.0
This major version introduces a new training algorithm that significantly reduces errors on noisy and multi-domain data, while maintaining previous v5 features and improving overall quality metrics.
Aug 26, 2025
- v5.3.0The Initialized Release
Focuses on optimization and initialization, adding a `Content.simplify` method and a new `Initialize` reactive for method-based initialization.
Aug 7, 2025
- v5.2.0The Streamed Layout
Adds a new experimental 'stream' layout, which is similar to vertical but faster due to fewer supported rules.
Aug 1, 2025
- v5.1.1The skinny release
A hotfix release that resolves an issue with PyPI package size due to pycache files being included in the build.
Jul 31, 2025
- v5.1.0The empty inside release
Adds a new `:empty` pseudo-class for matching widgets with no children and adds support for horizontal scrolling via trackpad.
Jul 31, 2025
- v5.0.1The appended release
A hotfix addressing an issue where appending to Markdown widgets constructed with existing documents would fail.
Jul 25, 2025
- v5.0.0The Tabled release.
A major release focusing on Markdown improvements including tables, text selection, and streaming support, plus the auto-import of speedups.
Jul 25, 2025
- 25.6.3
This is a dependency patch release that updates the pypdfium2 dependency from version 4.30.1 (which was yanked) to the latest stable version.
Jul 22, 2025
- v4.0.0The Streaming Release
Highlights Markdown streaming capabilities and a significant change to the `Widget.anchor` method semantics.
Jul 12, 2025