Releases
Version v4.1.0
Invoke v4.1.0 brings a many fixes and enhancements. The big ticket is Style and Composition IP Adapter.
🧪 Style and Composition IP Adapter (beta)
IP Adapter uses an image as a prompt. Images have two major components - their style and their composition - and you can choose either or both when using IP Adapter.
Use the new IP Adapter Method dropdown to select Full, Style, or Composition. The setting is applied per IP Adapter. You may need to delete and re-add active IP Adapters to see the dropdown.
| No IP Adapter | IP Adapter Image | Full IP Adapter | Style Only | Composition Only |
|---|---|---|---|---|
“a fierce wolf in an alpine forest”, all using same seed - note how the Full method turns the wolf into a mouse-canine hybrid
Shout-out to @blessedcoolant for this feature!
📈 Patch Nodes for v4.1.0
Enhancements
- Backend and nodes implementation for regional prompting and regional IP Adapter (UI in v4.2.0)
- Secret option in Workflow Editor to convert a graph into a workflow. See #6181 for how to use it.
- Assortment of UI papercuts
- Favicon & page title indicate generation status @jungleBadger
- Delete hotkey and button work with gallery selection @jungleBadger
- Workflow editor perf improvements
- Edge labels in workflow editor
- Updated translations @Harvester62, @symant233, @Vasyanator
- Updated docs @sarashinai
- Improved torch device and precision handling
Fixes
multipleOffor invocations (for example, the Noise invocation’s width and height have a step of 8)- Poor quality “fried” refiner outputs
- Poor quality inpainting with gradient denoising and refiner
- Canvas images appearing in the wrong places
- The little eye defaulting to off in canvas staging toolbar
- Premature OOM on windows (see shared GPU memory FAQ)
- ~1s delay between queue items
- Wonky model manager forms navigating away from UI @clsn
Invocation API
- New method to get the filesystem path of an image:
context.images.get_path(image_name: str, thumbnail: bool)@fieldOfView
Internal
- Improved
knipconfig @webpro - Updated python deps @Malrama
💾 Installation and Updating
To install or update to v4.1.0, download the installer and follow the installation instructions.
To update, select the same installation location. Your user data will not be touched.
Missing models after updating from v3 to v4
See this FAQ.
🐛 Known Issues
- Inpainting models on Canvas sometimes kinda give up and output mush. The fix didn’t make it in to v4.1.0, we will aim to release a patch in by the weekend.
What’s Changed
- fix(ui): handle multipleOf on number fields by @psychedelicious in https://github.com/invoke-ai/InvokeAI/pull/6167
- feat(backend): clean up choose_precision by @psychedelicious in https://github.com/invoke-ai/InvokeAI/pull/6171
- (feat) add a method to get the path of an image from invocationcontext by @fieldOfView in https://github.com/invoke-ai/InvokeAI/pull/6176
- feat(ui): add graph-to-workflow debug helper by @psychedelicious in https://github.com/invoke-ai/InvokeAI/pull/6181
- fix: Update SDXL Refiner graphs to use Gradient Mask by @blessedcoolant in https://github.com/invoke-ai/InvokeAI/pull/6158
- Regional Prompting (Node only) by @RyanJDick in https://github.com/invoke-ai/InvokeAI/pull/5916
- Remove Prompt-to-Prompt Cross-Attention Control by @RyanJDick in https://github.com/invoke-ai/InvokeAI/pull/6123
- Regional IP-Adapter (Backend Only) by @RyanJDick in https://github.com/invoke-ai/InvokeAI/pull/5967
- Fix field ordering by @JPPhoto in https://github.com/invoke-ai/InvokeAI/pull/6186
- Fix regional prompt mask padding by @RyanJDick in https://github.com/invoke-ai/InvokeAI/pull/6187
- fix(ui): always enqueue with fresh staging area bbox by @maryhipp in https://github.com/invoke-ai/InvokeAI/pull/6184
- fix(nodes): refined fixes for refiner inpainting by @psychedelicious in https://github.com/invoke-ai/InvokeAI/pull/6197
- [mm] clear the cache entry for a model that got an OOM during loading by @lstein in https://github.com/invoke-ai/InvokeAI/pull/6193
- fix(queue): poll queue on finished queue item by @psychedelicious in https://github.com/invoke-ai/InvokeAI/pull/6191
- Remove type=“submit” from all tsx files. by @clsn in https://github.com/invoke-ai/InvokeAI/pull/6200
- Add Toggle for showing Bounding Box, fix Main Model info popover, fix Control Model description on hover by @chainchompa in https://github.com/invoke-ai/InvokeAI/pull/6202
- Update INSTALL_DEVELOPMENT.md by @psychedelicious in https://github.com/invoke-ai/InvokeAI/pull/6201
- Correct stated location of Gallery panel in WEB.md by @sarashinai in https://github.com/invoke-ai/InvokeAI/pull/6203
- Fix “recall all” when loras, controls, or high res fix are not present by @chainchompa in https://github.com/invoke-ai/InvokeAI/pull/6211
- Modify title and favicon when item in progress (issue #2392) by @jungleBadger in https://github.com/invoke-ai/InvokeAI/pull/6204
- remove hires fix badge from settings when using sdxl by @chainchompa in https://github.com/invoke-ai/InvokeAI/pull/6218
- Enable multi deletion on keystroke by @jungleBadger in https://github.com/invoke-ai/InvokeAI/pull/6216
- ui: translations update from weblate by @weblate in https://github.com/invoke-ai/InvokeAI/pull/6170
- Fix up Knip config & enable —production mode by @webpro in https://github.com/invoke-ai/InvokeAI/pull/6217
- Update deps to their lastest versions by @Malrama in https://github.com/invoke-ai/InvokeAI/pull/6178
- feat(ui): edge labels by @psychedelicious in https://github.com/invoke-ai/InvokeAI/pull/6224
- perf(ui): misc ui improvements by @psychedelicious in https://github.com/invoke-ai/InvokeAI/pull/6225
- [util] Add generic torch device class by @lstein in https://github.com/invoke-ai/InvokeAI/pull/6174
- [mm] Fix race condition when installing multi-file models by @lstein in https://github.com/invoke-ai/InvokeAI/pull/6221
- Add IP Adapter Style & Composition Modes by @blessedcoolant in https://github.com/invoke-ai/InvokeAI/pull/6213
- update IP adapter method labels to be designated as Beta by @maryhipp in https://github.com/invoke-ai/InvokeAI/pull/6232
- ui: translations update from weblate by @weblate in https://github.com/invoke-ai/InvokeAI/pull/6223
- chore: v4.1.0 by @psychedelicious in https://github.com/invoke-ai/InvokeAI/pull/6234
New Contributors
- @sarashinai made their first contribution in https://github.com/invoke-ai/InvokeAI/pull/6203
- @jungleBadger made their first contribution in https://github.com/invoke-ai/InvokeAI/pull/6204
- @webpro made their first contribution in https://github.com/invoke-ai/InvokeAI/pull/6217
Full Changelog: https://github.com/invoke-ai/InvokeAI/compare/v4.0.4…v4.1.0
This site was designed and developed by Aether Fox Studio.