Releases: invoke-ai/InvokeAI
v6.7.0rc1
This minor release includes improved object selection on Canvas, layer adjustments, math evaluation in Workflow Editor number fields, and a handful of other enhancements and fixes.
Select Object v2
We've made some major improvements to object selection.
- Segment Anything v2 is now supported. You can choose between SAM1 and SAM2. We've found that SAM2 is much faster than SAM1, but often does not perform as well, so we left SAM1 as an option.
- You may now draw a box around the target object. The box doesn't need to be exact - sometimes, you can get better results by making it a bit smaller than the target object. Points are still supported and can be used independently or as a refinement for a box.
- Holding
shift
while clicking creates an exclude point if you have include selected. If you have selected exclude, holdingshift
will instead create an include point. - You can now provide a text prompt instead of a box and points. Use very simple language for best results. Internally, this uses Grounding DINO to identify the target.
Raster Layer Adjustments
Right click a Raster Layer to add adjustments. Adjustments are non-destructive, though you can accept them to bake them into the layer.
You can adjust brightness, contrast, saturation, temperature, tint, and sharpness, or use the curves editor to adjust each channel independently.
Thanks @dunkeroni for implementing this very useful feature.
Enhancements
- Improved object selection on Canvas.
- Raster layer adjustments. Thanks @dunkeroni!
- Support for mathematical expressions in number input fields. Currently, these are only enabled for fields in the Workflow Editor (including Builder Forms). Thanks @csehatt741!
- Queue list now sorts with newest on top. You can reverse the sort if you want, to restore the previous sorting. Thanks @csehatt741!
- Updated translations. Thanks @Harvester62 @Linos1391!
Fixes
- Fixed an issue that prevented you from using LoRA weights outside the range -1 to 2.
- Fixed a path traversal vulnerability affecting the bulk downloads API.
Installing and Updating
The Invoke Launcher is the recommended way to install, update and run Invoke. It takes care of a lot of details for you - like installing the right version of python - and runs Invoke as a desktop application.
Note: With recent updates to torch
, users on older GPUs (20xx and 10xx series) will likely run into issues with installing/updating. We are still evaluating how we can support older GPUs, but in the meantime users have found success manually downgrading torch
. Head over to discord if you need help.
Follow the Quick Start guide to get started with the launcher.
If you don't want to use the launcher, or need a headless install, you can follow the manual install guide.
What's Changed
- feat(ui): reverse queue list by @csehatt741 in #8488
- fix(ui): route metadata to gemini node by @psychedelicious in #8547
- fix(ui): LoRA number input min/max restored by @csehatt741 in #8542
- fix(app): path traversal via bulk downloads paths by @psychedelicious in #8548
- feat(ui): maths enabled on numeric input fields in workflow editor by @csehatt741 in #8549
- feat(ui): SAM2 Node & Integration by @hipsterusername in #8526
- queue list: remove completed_at, restore field values by @maryhipp in #8555
- ai(ui): add CLAUDE.md to frontend by @psychedelicious in #8556
- feat(ui): Raster Layer Color Adjusters by @dunkeroni in #8420
- ui: translations update from weblate by @weblate in #8545
- chore: prep for v6.7.0rc1 by @psychedelicious in #8557
Full Changelog: v6.6.0...v6.7.0rc1
v6.6.0
This is a minor release, adding a few QoL improvements and fixes.
Enhancements
- Canvas Color Picker has foreground and background colors. Switch between them with
x
. Pressd
to reset them to black and white. Thanks @csehatt741! - You can set a default weight setting for LoRAs in the Model Manager. When you add the LoRA, it will start at the default weight. Thanks @csehatt741!
- Canvas Brush/Eraser width renders an in-line slider when there is enough space instead of showing the slider in a popover.
- Updated translations. Thanks @Harvester62!
Fixes
- Always delete LoRAs when recalling all metadata. Thanks @csehatt741!
- Incompatible LoRAs being enabled prevents you from clicking Invoke.
- Fixed an issue where it was possible to drag a tab panel to another location in the UI on Chrome and Launcher (Firefox was unaffected).
- Internal file organization fix for docker builds.
- Fix an issue where progress images were super tiny (again).
- Fix an issue where no fallback was rendered in the viewer when no image is selected.
- Fix an issue where a single middle-mouse click on Canvas would activate the View tool (i.e. drag-to-pan), and you had to click again to deactivate it.
- Fix an issue in the Viewer where the last-generated image would briefly show after the current generation finishes.
Installing and Updating
The Invoke Launcher is the recommended way to install, update and run Invoke. It takes care of a lot of details for you - like installing the right version of python - and runs Invoke as a desktop application.
Note: With recent updates to torch
, users on older GPUs (20xx and 10xx series) will likely run into issues with installing/updating. We are still evaluating how we can support older GPUs, but in the meantime users have found success manually downgrading torch
. Head over to discord if you need help.
Follow the Quick Start guide to get started with the launcher.
If you don't want to use the launcher, or need a headless install, you can follow the manual install guide.
What's Changed
- video by @maryhipp in #8499
- do not show negative prompt or ref images on video tab by @maryhipp in #8501
- handle large videos by @maryhipp in #8503
- match screen capture button to the others by @maryhipp in #8504
- fix(app): board count queries not getting categories as params by @psychedelicious in #8505
- Add 'sd-2' to supported negative prompt base models by @damian0815 in #8513
- fix(ui): remove LoRAs for recall use all by @csehatt741 in #8512
- feat(ui): add readiness checks for LoRAs by @psychedelicious in #8506
- fix(ui): move
getItemsPerRow
to frontend src dir by @psychedelicious in #8515 - chore(ui): bump dockview by @psychedelicious in #8516
- ui: translations update from weblate by @weblate in #8507
- feat(ui): switchable foreground/background colors by @csehatt741 in #8510
- feat(ui): LoRA default weight by @csehatt741 in #8484
- ui(fix): remove video base models from image aspect/ratio logic by @maryhipp in #8521
- ui: translations update from weblate by @weblate in #8517
- chore: prep for v6.6.0rc1 by @psychedelicious in #8522
- fix(ui): fix situation where progress images are super tiny by @psychedelicious in #8524
- fix(ui): browser image caching cors race condition by @psychedelicious in #8523
- fix(ui): gallery selection issues by @psychedelicious in #8528
- fix(ui): stop dragging when user clicks mmb once by @psychedelicious in #8527
- fix(ui): prev image briefly showing in viewer as progress image "resolves" into output image by @psychedelicious in #8530
- chore: prep for v6.6.0rc2 by @psychedelicious in #8531
- tidy(ui): translation cleanup and CI checks by @psychedelicious in #8536
- tidy,fix(ui): remove unused coords from params slice by @psychedelicious in #8534
- ui: translations update from weblate by @weblate in #8533
- Revert "tidy(ui): translation cleanup and CI checks" by @psychedelicious in #8539
- feat(ui): slider for brush and eraser tool by @csehatt741 in #8525
- chore: prep for v6.6.0 by @psychedelicious in #8538
Full Changelog: v6.5.1...v6.6.0
v6.6.0rc2
This is a minor release, adding a few QoL improvements and fixes.
Enhancements
- Canvas Color Picker has foreground and background colors. Switch between them with
x
. Pressd
to reset them to black and white. Thanks @csehatt741! - You can set a default weight setting for LoRAs in the Model Manager. When you add the LoRA, it will start at the default weight. Thanks @csehatt741!
- Updated translations. Thanks @Harvester62!
Fixes
- Always delete LoRAs when recalling all metadata. Thanks @csehatt741!
- Incompatible LoRAs being enabled prevents you from clicking Invoke.
- Fixed an issue where it was possible to drag a tab panel to another location in the UI on Chrome and Launcher (Firefox was unaffected).
- Internal file organization fix for docker builds.
- Fix an issue where progress images were super tiny (again).
- Fix an issue where no fallback was rendered in the viewer when no image is selected.
- Fix an issue where a single middle-mouse click on Canvas would activate the View tool (i.e. drag-to-pan), and you had to click again to deactivate it.
- Fix an issue in the Viewer where the last-generated image would briefly show after the current generation finishes.
Installing and Updating
The Invoke Launcher is the recommended way to install, update and run Invoke. It takes care of a lot of details for you - like installing the right version of python - and runs Invoke as a desktop application.
Note: With recent updates to torch
, users on older GPUs (20xx and 10xx series) will likely run into issues with installing/updating. We are still evaluating how we can support older GPUs, but in the meantime users have found success manually downgrading torch
. Head over to discord if you need help.
Follow the Quick Start guide to get started with the launcher.
If you don't want to use the launcher, or need a headless install, you can follow the manual install guide.
What's Changed
- video by @maryhipp in #8499
- do not show negative prompt or ref images on video tab by @maryhipp in #8501
- handle large videos by @maryhipp in #8503
- match screen capture button to the others by @maryhipp in #8504
- fix(app): board count queries not getting categories as params by @psychedelicious in #8505
- Add 'sd-2' to supported negative prompt base models by @damian0815 in #8513
- fix(ui): remove LoRAs for recall use all by @csehatt741 in #8512
- feat(ui): add readiness checks for LoRAs by @psychedelicious in #8506
- fix(ui): move
getItemsPerRow
to frontend src dir by @psychedelicious in #8515 - chore(ui): bump dockview by @psychedelicious in #8516
- ui: translations update from weblate by @weblate in #8507
- feat(ui): switchable foreground/background colors by @csehatt741 in #8510
- feat(ui): LoRA default weight by @csehatt741 in #8484
- ui(fix): remove video base models from image aspect/ratio logic by @maryhipp in #8521
- ui: translations update from weblate by @weblate in #8517
- chore: prep for v6.6.0rc1 by @psychedelicious in #8522
- fix(ui): fix situation where progress images are super tiny by @psychedelicious in #8524
- fix(ui): browser image caching cors race condition by @psychedelicious in #8523
- fix(ui): gallery selection issues by @psychedelicious in #8528
- fix(ui): stop dragging when user clicks mmb once by @psychedelicious in #8527
- fix(ui): prev image briefly showing in viewer as progress image "resolves" into output image by @psychedelicious in #8530
- chore: prep for v6.6.0rc2 by @psychedelicious in #8531
Full Changelog: v6.5.1...v6.6.0rc2
v6.6.0rc1
This is a minor release, adding a few QoL improvements and fixes.
Enhancements
- Canvas Color Picker has foreground and background colors. Switch between them with
x
. Pressd
to reset them to black and white. Thanks @csehatt741! - You can set a default weight setting for LoRAs in the Model Manager. When you add the LoRA, it will start at the default weight. Thanks @csehatt741!
- Updated translations. Thanks @Harvester62!
Fixes
- Always delete LoRAs when recalling all metadata. Thanks @csehatt741!
- Incompatible LoRAs being enabled prevents you from clicking Invoke.
- Fixed an issue where it was possible to drag a tab panel to another location in the UI on Chrome and Launcher (Firefox was unaffected).
- Internal file organization fix for docker builds.
Installing and Updating
The Invoke Launcher is the recommended way to install, update and run Invoke. It takes care of a lot of details for you - like installing the right version of python - and runs Invoke as a desktop application.
Note: With recent updates to torch
, users on older GPUs (20xx and 10xx series) will likely run into issues with installing/updating. We are still evaluating how we can support older GPUs, but in the meantime users have found success manually downgrading torch
. Head over to discord if you need help.
Follow the Quick Start guide to get started with the launcher.
If you don't want to use the launcher, or need a headless install, you can follow the manual install guide.
What's Changed
- video by @maryhipp in #8499
- do not show negative prompt or ref images on video tab by @maryhipp in #8501
- handle large videos by @maryhipp in #8503
- match screen capture button to the others by @maryhipp in #8504
- fix(app): board count queries not getting categories as params by @psychedelicious in #8505
- Add 'sd-2' to supported negative prompt base models by @damian0815 in #8513
- fix(ui): remove LoRAs for recall use all by @csehatt741 in #8512
- feat(ui): add readiness checks for LoRAs by @psychedelicious in #8506
- fix(ui): move
getItemsPerRow
to frontend src dir by @psychedelicious in #8515 - chore(ui): bump dockview by @psychedelicious in #8516
- ui: translations update from weblate by @weblate in #8507
- feat(ui): switchable foreground/background colors by @csehatt741 in #8510
- feat(ui): LoRA default weight by @csehatt741 in #8484
- ui(fix): remove video base models from image aspect/ratio logic by @maryhipp in #8521
- ui: translations update from weblate by @weblate in #8517
- chore: prep for v6.6.0rc1 by @psychedelicious in #8522
Full Changelog: v6.5.1...v6.6.0rc1
v6.5.1
This is a patch release, fixing a few high priority bugs.
Fixes
- Hard crash when generating with FLUX on Windows.
- Super tiny progress images on Canvas.
- Assorted Canvas issues, mostly around transparency.
Installing and Updating
The Invoke Launcher is the recommended way to install, update and run Invoke. It takes care of a lot of details for you - like installing the right version of python - and runs Invoke as a desktop application.
Note: With recent updates to torch
, users on older GPUs (20xx and 10xx series) will likely run into issues with installing/updating. We are still evaluating how we can support older GPUs, but in the meantime users have found success manually downgrading torch
. Head over to discord if you need help.
Follow the Quick Start guide to get started with the launcher.
If you don't want to use the launcher, or need a headless install, you can follow the manual install guide.
What's Changed
- fix(ui): control layer transparency effect not working by @psychedelicious in #8496
- fix(ui): progress image renders at physical size by @psychedelicious in #8495
- ui: translations update from weblate by @weblate in #8498
- fix(app): FLUX on Windows hard crash by @psychedelicious in #8494
Full Changelog: v6.5.0...v6.5.1
v6.5.0
This release includes a handful of fixes and enhancements.
Enhancements
- Add a optional
Shuffle
button to float and integer fields in Workflow Builder forms. Thanks @csehatt741! - Canvas color picker non longer changes the alpha of the color.
- When the bbox aspect ratio is locked, resizing the bbox from the Canvas will respect the locked status of the aspect ratio. Hold
shift
to temporarily invert the locked status:- When the aspect ratio is locked, holding
shift
while resizing the bbox will allow you to freely resize the bbox. - When the aspect ratio is not locked, holding
shift
while resizing the bbox will maintain the last aspect ratio of the bbox.
- When the aspect ratio is locked, holding
- When a node field is added to a Workflow Builder form, the
+
button to add it will now show a-
and let you remove the field. Thanks @csehatt741! - When changing a selection of image's board, the current board is hidden from the board drop-down. The items in the drop down are now sorted alphabetically. Thanks @csehatt741!
- When using a model that doesn't support reference images, they will be hidden. You can now Invoke without needing to disable them.
- When using a model that doesn't support explicit width and height settings, they will be hidden.
Fixes
- Rare issue with HF tokens that could cause an error when downloading models from a protect HF repo immediately after setting the token in Invoke's Model Manager.
- Fix an issue with float field precision in the Workflow editor.
- Fix an error
AttributeError: module 'cv2.ximgproc' has no attribute 'thinning'
. Affected users should use the Launcher's Repair Mode to get the fix, otherwise the error will persist. - Disable the color picker when using middle mouse to pan the Canvas.
- Minor issue related to gallery multi-select where the last-selected image didn't show in the viewer.
- Prevent dragging and dropping a node field into the Workflow Builder if it has already been added once.
- Fix an issue where the last progress image for a Canvas generation would get stuck on the Viewer tab.
- Fix an issue where certain image loading errors in Canvas were not logged correctly.
Installing and Updating
The Invoke Launcher is the recommended way to install, update and run Invoke. It takes care of a lot of details for you - like installing the right version of python - and runs Invoke as a desktop application.
Note: With recent updates to torch
, users on older GPUs (20xx and 10xx series) will likely run into issues with installing/updating. We are still evaluating how we can support older GPUs, but in the meantime users have found success manually downgrading torch
. Head over to discord if you need help.
Follow the Quick Start guide to get started with the launcher.
If you don't want to use the launcher, or need a headless install, you can follow the manual install guide.
What's Changed
- chore: prep for v6.4.0 by @psychedelicious in #8447
- ci: add workflow to catch incorrect usage of git-lfs by @psychedelicious in #8452
- feat(ui): shuffle button on workflows by @csehatt741 in #8450
- docs: update quick start instructions & links by @psychedelicious in #8454
- feat(ui): do not sample alpha in Canvas color picker by @psychedelicious in #8448
- fix(ui): race condition when setting hf token and downloading model by @psychedelicious in #8456
- fix(ui): float input precision by @psychedelicious in #8455
- feat(app): vendor in
invisible-watermark
by @psychedelicious in #8468 - fix(ui): disable color picker while middle-mouse panning canvas by @psychedelicious in #8475
- fix(ui): toggle bbox visiblity translation by @psychedelicious in #8476
- feat(ui): respect aspect ratio when resizing bbox on canvas by @psychedelicious in #8474
- feat(ui): remove input field from form button on node field by @csehatt741 in #8457
- fix(ui): respect direction of selection in Gallery by @csehatt741 in #8458
- ui: translations update from weblate by @weblate in #8464
- chore: prep for v6.5.0rc1 by @psychedelicious in #8479
- feat(ui): bbox aspect ratio lock is always inverted by shift by @psychedelicious in #8477
- feat(ui): change board - sorting order of boards alphabetical by @csehatt741 in #8460
- Feat/same field multiple times added to form by @csehatt741 in #8480
- UI support for gemini 2.5 flash image by @maryhipp in #8489
- update copy for API models without w/h controls by @maryhipp in #8490
- fix(ui): progress image gets stuck on viewer when generating on canvas by @psychedelicious in #8483
- fix(ui): konva logging by @psychedelicious in #8493
- ui: translations update from weblate by @weblate in #8485
- chore: prep for v6.5.0 by @psychedelicious in #8492
Full Changelog: v6.4.0...v6.5.0
v6.5.0rc1
This release includes a handful of fixes and enhancements.
Enhancements
- Add a optional
Shuffle
button to float and integer fields in Workflow Builder forms. Thanks @csehatt741! - Canvas color picker non longer changes the alpha of the color.
- When the bbox aspect ratio is locked, resizing the bbox from the Canvas will respect the locked status of the aspect ratio. Hold
shift
to temporarily invert the locked status:- When the aspect ratio is locked, holding
shift
while resizing the bbox will allow you to freely resize the bbox. - When the aspect ratio is not locked, holding
shift
while resizing the bbox will maintain the last aspect ratio of the bbox.
- When the aspect ratio is locked, holding
- When a node field is added to a Workflow Builder form, the
+
button to add it will now show a-
and let you remove the field.
Fixes
- Rare issue with HF tokens that could cause an error when downloading models from a protect HF repo immediately after setting the token in Invoke's Model Manager.
- Fix an issue with float field precision in the Workflow editor.
- Fix an error
AttributeError: module 'cv2.ximgproc' has no attribute 'thinning'
. Affected users should use the Launcher's Repair Mode to get the fix, otherwise the error will persist. - Disable the color picker when using middle mouse to pan the Canvas.
- Minor issue related to gallery multi-select where the last-selected image didn't show in the viewer.
Installing and Updating
The Invoke Launcher is the recommended way to install, update and run Invoke. It takes care of a lot of details for you - like installing the right version of python - and runs Invoke as a desktop application.
Note: With recent updates to torch
, users on older GPUs (20xx and 10xx series) will likely run into issues with installing/updating. We are still evaluating how we can support older GPUs, but in the meantime users have found success manually downgrading torch
. Head over to discord if you need help.
Follow the Quick Start guide to get started with the launcher.
If you don't want to use the launcher, or need a headless install, you can follow the manual install guide.
What's Changed
- ui: translations update from weblate by @weblate in #8432
- git: move test LoRA to LFS by @psychedelicious in #8446
- chore: prep for v6.4.0 by @psychedelicious in #8447
- ci: add workflow to catch incorrect usage of git-lfs by @psychedelicious in #8452
- feat(ui): shuffle button on workflows by @csehatt741 in #8450
- docs: update quick start instructions & links by @psychedelicious in #8454
- feat(ui): do not sample alpha in Canvas color picker by @psychedelicious in #8448
- fix(ui): race condition when setting hf token and downloading model by @psychedelicious in #8456
- fix(ui): float input precision by @psychedelicious in #8455
- feat(app): vendor in
invisible-watermark
by @psychedelicious in #8468 - fix(ui): disable color picker while middle-mouse panning canvas by @psychedelicious in #8475
- fix(ui): toggle bbox visiblity translation by @psychedelicious in #8476
- feat(ui): respect aspect ratio when resizing bbox on canvas by @psychedelicious in #8474
- feat(ui): remove input field from form button on node field by @csehatt741 in #8457
- fix(ui): respect direction of selection in Gallery by @csehatt741 in #8458
- ui: translations update from weblate by @weblate in #8464
- feat(ui): bbox aspect ratio lock is always inverted by shift by @psychedelicious in #8477
Full Changelog: v6.4.0rc2...v6.5.0rc1
v6.4.0
This release includes a handful of fixes and enhancements.
Enhancements
Shout-out to @csehatt741 for knocking out some great QoL improvements. Thank you!
- Canvas Bbox visibility can be toggled with
shift+o
. Thanks @csehatt741! - Nodes with execution errors are highlighted red. Thanks @csehatt741!
- Prevent a field from being added to Workflow Builder forms multiple times. Thanks @csehatt741!
- Support recall of CLIP Skip metadata. Thanks @csehatt741!
- Fixed some issues with model install paths.
- Tweaked state persistence strategy - now debounced to 300ms instead of throttled to 2000ms. This should reduce stutters while doing things like panning around the Canvas.
- SDXL Style prompts have been removed from the Generate, Canvas and Upscaling tabs. This rarely-used setting was unintuitive at best. You can still use it in Workflows, but we are removing this footgun from the linear UI tabs.
- Prompt and seed metadata may now be recalled on the Upscaling tab.
- The buttons to download potentially very large starter model bundles show a confirmation dialog before starting the download. Thanks @csehatt741!
- Merged layers are inserted in the right spot in the layers panel. Thanks @csehatt741!
- Added button to image context menu and viewer toolbar to locate an image in the gallery. The image's board is selected and image scrolled to. Thanks @csehatt741!
- Support FLUX PEFT LoRAs with
base_model.model
key prefix. - Improved VAE encode VRAM usage.
- Updated translations. Thanks @Harvester62!
Fixes
- Minor bug when concatenating Kontext ref images in latent space that could result in some images not being "seen".
- Fit to Bbox functionality could result in the layer being sized correctly but positioned incorrectly when the bbox was not aligned to the 64px grid.
- Allow use of mouse in node title editable inputs.
Internal/Dev
- Fix AMD docker image build issue related to disk space. Thanks @heathen711!
Installing and Updating
The Invoke Launcher is the recommended way to install, update and run Invoke. It takes care of a lot of details for you - like installing the right version of python - and runs Invoke as a desktop application.
Note: With recent updates to torch
, users on older GPUs (20xx and 10xx series) will likely run into issues with installing/updating. We are still evaluating how we can support older GPUs, but in the meantime users have found success manually downgrading torch
. Head over to discord if you need help.
Follow the Quick Start guide to get started with the launcher.
If you don't want to use the launcher, or need a headless install, you can follow the manual install guide.
What's Changed
- chore: prep for v6.3.0 by @psychedelicious in #8397
- bugfix(container-builder) Use the mnt space instead of root space for docker images by @heathen711 in #8361
- feat(ui): add toggle for bbox with hotkey by @csehatt741 in #8385
- feat(ui): outline error nodes in red by @csehatt741 in #8398
- fix(ui): same field cannot be added to form multiple times in workflow editor by @csehatt741 in #8403
- feat(mm): improved VAE encode VRAM usage by @psychedelicious in #8414
- fix(mm): only add suffix to model paths when path is file by @psychedelicious in #8413
- feat(ui): debounce persistence instead of throttle by @psychedelicious in #8410
- fix(ui): upscaling prompt metadata by @psychedelicious in #8411
- fix(ui): input field error styling specificity by @psychedelicious in #8418
- feat(ui): add missing translation strings by @psychedelicious in #8412
- fix(mm): fail when model exists at path instead of finding unused new path by @psychedelicious in #8419
- Update NODES.md by @solkyoshiro in #8378
- chore: fix some comments by @jiangmencity in #7575
- feat(ui): remove SDXL style prompt from linear UI by @psychedelicious in #8417
- ui: translations update from weblate by @weblate in #8404
- feat(ui)/clip skip by @csehatt741 in #8422
- feat(ui): confirmation before downloading starter bundle by @csehatt741 in #8427
- feat(ui): layer behaviour after merging by @csehatt741 in #8428
- feat(ui): locate in gallery image context menu by @csehatt741 in #8434
- tests: skip flaky MPS tests on CI by @psychedelicious in #8442
- fix(ui): update board totals when generation completes by @psychedelicious in #8424
- fix(ui): export NumericalParameterConfig type by @psychedelicious in #8436
- refactor: estimate working vae memory during encode/decode by @psychedelicious in #8438
- Support PEFT Loras with Base_Model.model prefix by @hipsterusername in #8433
- fix(ui): fit to bbox when bbox is not aligned to 64px grid by @psychedelicious in #8439
- fix(ui): prevent node drag when editing title by @psychedelicious in #8440
- chore: prep for v6.4.0rc2 by @psychedelicious in #8441
- ui: translations update from weblate by @weblate in #8432
- git: move test LoRA to LFS by @psychedelicious in #8446
New Contributors
- @csehatt741 made their first contribution in #8385
- @solkyoshiro made their first contribution in #8378
- @jiangmencity made their first contribution in #7575
Full Changelog: v6.3.0...v6.4.0
v6.4.0rc2
This minor release includes a handful of fixes and enhancements.
Enhancements
Shout-out to @csehatt741 for knocking out some great QoL improvements. Thank you!
- Canvas Bbox visibility can be toggled with
shift+o
. Thanks @csehatt741! - Nodes with execution errors are highlighted red. Thanks @csehatt741!
- Prevent a field from being added to Workflow Builder forms multiple times. Thanks @csehatt741!
- Support recall of CLIP Skip metadata. Thanks @csehatt741!
- Fixed some issues with model install paths.
- Tweaked state persistence strategy - now debounced to 300ms instead of throttled to 2000ms. This should reduce stutters while doing things like panning around the Canvas.
- SDXL Style prompts have been removed from the Generate, Canvas and Upscaling tabs. This rarely-used setting was unintuitive at best. You can still use it in Workflows, but we are removing this footgun from the linear UI tabs.
- Prompt and seed metadata may now be recalled on the Upscaling tab.
- The buttons to download potentially very large starter model bundles show a confirmation dialog before starting the download. Thanks @csehatt741!
- Merged layers are inserted in the right spot in the layers panel. Thanks @csehatt741!
- Added button to image context menu and viewer toolbar to locate an image in the gallery. The image's board is selected and image scrolled to. Thanks @csehatt741!
- Support FLUX PEFT LoRAs with
base_model.model
key prefix. - Updated translations. Thanks @Harvester62!
Fixes
- Improved VAE encode VRAM usage.
- Minor bug when concatenating Kontext ref images in latent space that could result in some images not being "seen".
- Fit to Bbox functionality could result in the layer being sized correctly but positioned incorrectly when the bbox was not aligned to the 64px grid.
- Allow use of mouse in node title editable inputs.
Internal/Dev
- Fix AMD docker image build issue related to disk space. Thanks @heathen711!
Installing and Updating
The Invoke Launcher is the recommended way to install, update and run Invoke. It takes care of a lot of details for you - like installing the right version of python - and runs Invoke as a desktop application.
Note: With recent updates to torch
, users on older GPUs (20xx and 10xx series) will likely run into issues with installing/updating. We are still evaluating how we can support older GPUs, but in the meantime users have found success manually downgrading torch
. Head over to discord if you need help.
Follow the Quick Start guide to get started with the launcher.
If you don't want to use the launcher, or need a headless install, you can follow the manual install guide.
What's Changed
- chore: prep for v6.3.0 by @psychedelicious in #8397
- bugfix(container-builder) Use the mnt space instead of root space for docker images by @heathen711 in #8361
- feat(ui): add toggle for bbox with hotkey by @csehatt741 in #8385
- feat(ui): outline error nodes in red by @csehatt741 in #8398
- fix(ui): same field cannot be added to form multiple times in workflow editor by @csehatt741 in #8403
- feat(mm): improved VAE encode VRAM usage by @psychedelicious in #8414
- fix(mm): only add suffix to model paths when path is file by @psychedelicious in #8413
- feat(ui): debounce persistence instead of throttle by @psychedelicious in #8410
- fix(ui): upscaling prompt metadata by @psychedelicious in #8411
- fix(ui): input field error styling specificity by @psychedelicious in #8418
- feat(ui): add missing translation strings by @psychedelicious in #8412
- fix(mm): fail when model exists at path instead of finding unused new path by @psychedelicious in #8419
- Update NODES.md by @solkyoshiro in #8378
- chore: fix some comments by @jiangmencity in #7575
- feat(ui): remove SDXL style prompt from linear UI by @psychedelicious in #8417
- ui: translations update from weblate by @weblate in #8404
- feat(ui)/clip skip by @csehatt741 in #8422
- feat(ui): confirmation before downloading starter bundle by @csehatt741 in #8427
- feat(ui): layer behaviour after merging by @csehatt741 in #8428
- feat(ui): locate in gallery image context menu by @csehatt741 in #8434
- tests: skip flaky MPS tests on CI by @psychedelicious in #8442
- fix(ui): update board totals when generation completes by @psychedelicious in #8424
- fix(ui): export NumericalParameterConfig type by @psychedelicious in #8436
- refactor: estimate working vae memory during encode/decode by @psychedelicious in #8438
- Support PEFT Loras with Base_Model.model prefix by @hipsterusername in #8433
- fix(ui): fit to bbox when bbox is not aligned to 64px grid by @psychedelicious in #8439
- fix(ui): prevent node drag when editing title by @psychedelicious in #8440
- chore: prep for v6.4.0rc2 by @psychedelicious in #8441
New Contributors
- @csehatt741 made their first contribution in #8385
- @solkyoshiro made their first contribution in #8378
- @jiangmencity made their first contribution in #7575
Full Changelog: v6.3.0...v6.4.0rc2
v6.4.0rc1
This minor release includes a handful of fixes and enhancements.
Enhancements
- Canvas Bbox visibility can be toggled with
shift+o
. Thanks @csehatt741! - Nodes with execution errors are highlighted red. Thanks @csehatt741!
- Prevent a field from being added to Workflow Builder forms multiple times. Thanks @csehatt741!
- Support recall of CLIP Skip metadata. Thanks @csehatt741!
- Fixed some issues with model install paths.
- Tweaked state persistence strategy - now debounced to 300ms instead of throttled to 2000ms. This should reduce stutters while doing things like panning around the Canvas.
- SDXL Style prompts have been removed from the Generate, Canvas and Upscaling tabs. This rarely-used setting was unintuitive at best. You can still use it in Workflows, but we are removing this footgun from the linear UI tabs.
- Prompt and seed metadata may now be recalled on the Upscaling tab.
- Updated translations. Thanks @Harvester62!
Fixes
- Improved VAE encode VRAM usage.
- Minor bug when concatenating Kontext ref images in latent space that could result in some images not being "seen".
Internal/Dev
- Fix AMD docker image build issue related to disk space. Thanks @heathen711!
Installing and Updating
The Invoke Launcher is the recommended way to install, update and run Invoke. It takes care of a lot of details for you - like installing the right version of python - and runs Invoke as a desktop application.
Note: With recent updates to torch
, users on older GPUs (20xx and 10xx series) will likely run into issues with installing/updating. We are still evaluating how we can support older GPUs, but in the meantime users have found success manually downgrading torch
. Head over to discord if you need help.
Follow the Quick Start guide to get started with the launcher.
If you don't want to use the launcher, or need a headless install, you can follow the manual install guide.
What's Changed
- chore: prep for v6.3.0 by @psychedelicious in #8397
- bugfix(container-builder) Use the mnt space instead of root space for docker images by @heathen711 in #8361
- feat(ui): add toggle for bbox with hotkey by @csehatt741 in #8385
- feat(ui): outline error nodes in red by @csehatt741 in #8398
- fix(ui): same field cannot be added to form multiple times in workflow editor by @csehatt741 in #8403
- feat(mm): improved VAE encode VRAM usage by @psychedelicious in #8414
- fix(mm): only add suffix to model paths when path is file by @psychedelicious in #8413
- feat(ui): debounce persistence instead of throttle by @psychedelicious in #8410
- fix(ui): upscaling prompt metadata by @psychedelicious in #8411
- fix(ui): input field error styling specificity by @psychedelicious in #8418
- feat(ui): add missing translation strings by @psychedelicious in #8412
- fix(mm): fail when model exists at path instead of finding unused new path by @psychedelicious in #8419
- Update NODES.md by @solkyoshiro in #8378
- chore: fix some comments by @jiangmencity in #7575
- feat(ui): remove SDXL style prompt from linear UI by @psychedelicious in #8417
- ui: translations update from weblate by @weblate in #8404
- feat(ui)/clip skip by @csehatt741 in #8422
New Contributors
- @csehatt741 made their first contribution in #8385
- @solkyoshiro made their first contribution in #8378
- @jiangmencity made their first contribution in #7575
Full Changelog: v6.3.0...v6.4.0rc1