Alpha v0.25.0
The first stable release in over a year. Two iterators can now be used together, as long as they are the same length.
Release announcements. These get written when a release needs more explaining than its changelog gives it.
The first stable release in over a year. Two iterators can now be used together, as long as they are the same length.
Conditional branching, a skipped state for nodes, selectable themes, and PyTorch upscaling roughly 1.25x faster than before.
Searchable documentation for every node, a rewritten update screen, backend setup changes, and the first Rust code in chaiNNer.
A year on from the first public release — what's planned for the year ahead, a website, and the beginnings of a model database.
A much more compact node design, inserting a node onto an existing connection, explanations for invalid connections, sharper previews, and DDS support on Windows.
Video support rebuilt on a full FFmpeg build, and auto-tiling reworked to use real tile sizes instead of a split factor.
GFPGAN face restoration arrives as the Face Upscale node, and the repository moves to the chaiNNer-org organization.
Direct PyTorch to NCNN conversion, SwinIR support, faster PyTorch auto-tiling, and lower RAM overhead on large chains.
Image previews inside the editor, via a new View Image node and a rework of how previews and starting nodes work.
Nodes already on the canvas can be dragged into an iterator, and released back out again.
The Navi type system, colored handles that show you what connects to what, a node search context menu, and the ability to disable nodes.
Batch upscaling arrives, through a new kind of node called an iterator, along with DDS and TGA support via Pillow.
NCNN support brings upscaling to AMD and Intel GPUs, and every node gets renamed and recategorized around what it does rather than what implements it.
The initial alpha release, and the first day of chaiNNer being public.