Releases: mashb1t/Fooocus
Releases · mashb1t/Fooocus
v2.5.2
What's Changed
- fix: add missing prompt return values for upscale fast 2x in enhance
- fix: add handling for only enhance upscale fast to not divide by 0 on step calculation
- fix: get upscale model filepath by calling downloading_upscale_model() to ensure the model exists
- fix: correctly break after user stopped during enhance inpainting
Full Changelog: v2.5.1...v2.5.2
v2.5.1
What's Changed
- fix: add missing handling for cloth category for u2net_cloth_seg
- fix: move current_task_id increase to handler again
- refactor: rename checkbox
Enable Mask Upload
toEnable Advanced Masking Features
Full Changelog: v2.5.0...v2.5.1
v2.5.0
What's Changed
- feat: add enhance feature, which offers easy image refinement steps (similar to adetailer, but based on dynamic image detection instead of specific mask detection models). See documentation.
- feat: improve GroundingDINO and SAM image masking
- refactor: rewrite async worker code, make code much more reusable to allow iterations and improve reusability
- refactor: rename
--enable-describe-uov-image
to--enable-auto-describe-image
, now also works for enhance image upload - feat: update python dependencies, remove diffusers, add segment_anything
- refactor: move checkboxes
Enable Mask Upload
andInvert Mask When Generating
fromDeveloper Debug Mode
toInpaint Or Outpaint
Full Changelog: v2.4.3...v2.5.0
v2.5.0-rc1
What's Changed
- feat: add enhance feature, which offers easy image refinement steps (similar to adetailer, but based on dynamic image detection instead of specific mask detection models). See lllyasviel#3122 for more details.
- feat: improve GroundingDINO and SAM image masking
Full Changelog: v2.4.3...v2.5.0-rc1
v2.4.3
What's Changed
- fix: correctly set alphas_cumprod by @mashb1t in lllyasviel#3106
- feat: parse env var strings to expected config value types by @mashb1t in lllyasviel#3107
Full Changelog: v2.4.2...v2.4.3
v2.4.2
What's Changed (synched from upstream)
- feat: add support and preset for playground v2.5 (only works with performance Quality or Speed, use with scheduler edm_playground_v2) by @mashb1t in lllyasviel#3073
- feat: make textboxes (incl. positive prompt) resizable by @mashb1t in lllyasviel#3074
- fix: use default vae name instead of None on file refresh by @mashb1t in lllyasviel#3045
- fix: correctly use translation and dynamic label for aspect ratios by @mashb1t in lllyasviel#3046
- feat: optimize performance lora filtering in metadata by @mashb1t in lllyasviel#3048
- feat: rework intermediate image display for restricted performances by @mashb1t in lllyasviel#3050
- fix: turbo scheduler loading issue by @mashb1t in lllyasviel#3065
- fix: chown files directly at copy in Dockerfile by @mashb1t in lllyasviel#3066
- feat: sync cmd args in readme by @mashb1t in lllyasviel#3075
- fix: correct sampling for tcd scheduler when gamma is 0 by @mashb1t in lllyasviel#3093
Full Changelog: lllyasviel/Fooocus@v2.4.1...v2.4.2
v2.4.1
What's Changed (synched from upstream)
- fix: adjust clip skip default value from 1 to 2 by @mashb1t in lllyasviel#3011
- fix: add type check for undefined, use fallback when no translation for aspect ratios was given by @mashb1t in lllyasviel#3025
- feat: build docker image tagged "edge" on push to main branch by @mashb1t in lllyasviel#3026
Full Changelog: lllyasviel/Fooocus@v2.4.0...v2.4.1
v2.4.0
All of 2.4.0 and a bit more.
You can find the docker images in https://ghcr.io/mashb1t/fooocus
Full Changelog: 2.3.1...v2.4.0
2.4.0-rc3
2.3.3
What's Changed
- feat: support download of huggingface files from a mirror website
- feat: update interposer from v3.1 to v4.0
- feat: update animaPencilXL from 2.0.0 to 3.1.0
- feat: add button to reconnect UI without having to reload the page
- feat: add performance hyper-sd based on 4step LoRA
- feat: add random style checkbox
- feat: add VAE select
- feat: add full raw prompt to history log
- feat: add AlignYourStepsScheduler
- feat: add tcd sampler
- refactor: code cleanup
Full Changelog: 2.3.1...2.3.3