- Fooocus to v2.5.3
- Add enhance image endpoint
- Add generate mask endpoint
- Influenced by Fooocus, the worker.py was reconstructed
- Update docs
- Returned base64 str now include identifier like this
data:image/jpeg;base64,
- Issue #375
- Issue #378
- Save extension params now also takes effect for base64 str in the returned data
- LoraManager support tar file
- Update Fooocus to v2.4.3
- Issue #363
- Add nsfw checker
- Fooocus to v2.4.1
- Increase support for config.txt
- Issue #319
- Issue #327
- Issue #332
- Fix JPEG support
- UA in image request
- Fooocus to v2.3.1
- Code formatting.
- Issue #302
- Issue #294
- Sync v1 endpoints.
- meta_scheme error
- Image meta save to image
- Code formatting.
- Fix opencv-python-headless failed in Docker
- Issue #270
- Issue #280
- Issue #222
- Url support for lora in replicate
- Update docs
- Rewrite Dockerfile
- Rewirte examples
- Full code of Fooocus include
- Remove related code
- Optimize project structure.
- Support for Lightning model.
- Update default checkpoint for Replicate.
- Issue #244
- Issue #259
- Issue #232
- Support for Fooocus 2.3.0.
- Project structure optimization.
- Removed unnecessary dependencies and optimized code.
- Save extension
- Secure API future with the use of API keys.
- Add seeds for predict output
- Update Docs
- OOM when running a long time
- Some spell error
- Issue with the Fooocus-API version in startup output.
- Support url for input_image in v2 API.
- Image Prompt Mixing requirements implemented
- Add SQLite database support for history.
- Update Docs
- Large queue size support
- Optimized async task response when the queue is full
- Update cog branch
- Optimized cli flages parser.
- Optimized some code formatting.
- Optimized the underlying logic of task execution.
- Default queue history size to 0 for no limit.
- Fix condition. default params broke here and this fixes auto mixing feature.
- Fix error when use `Extreme Speed' with cog.
- Fix typo of 'presistent'
- Image Prompt Mixing requirements implemented
- Some spell error, some translations.
- Fix image prompt must require 'input_image'.
- Implemented support for storing history to the database.
- Add example using ipynb
- Add error logging
- Add check for aspect_ratios_selection
- Image Prompt Mixing requirements implemented.
- Update Docs
- Merge Fooocus to v2.1.860
- Various bugs and issues reported by the community.
- Add ping endpoint
- Describe interface to get prompts from images.
- Add image_prompt to text2img endpoint
- Add mirror for fooocus
- Update Docs
- Added query job history API and webhook_url support for each generation request.
- Change to exit when Fooocus check failed
- Fix #122 query job not found error.
Please note that this ChangeLog is a summary and may not include all changes. For a complete list of changes, please refer to the commit history on the Fooocus-API GitHub repository.