- 🎉 video generation using [Stable Video Diffusion](https://github.com/Stability-AI/generative-models)
- add `--videogen` to any image generation to create a short video from the generated image
- or use `aimg videogen` to generate a video from an image
- 🎉 SDXL (Stable Diffusion Extra Large) models are now supported.
- try `--model opendalle` or `--model sdxl`
- inpainting and controlnets are not yet supported for SDXL
- 🎉 imaginairy is now backed by the [refiners library](https://github.com/finegrain-ai/refiners)
- This was a huge rewrite which is why some features are not yet supported. On the plus side, refiners supports
cutting edge features (SDXL, image prompts, etc) which will be added to imaginairy soon.
- [self-attention guidance](https://github.com/SusungHong/Self-Attention-Guidance) which makes details of images more accurate
- 🎉 feature: larger image generations now work MUCH better and stay faithful to the same image as it looks at a smaller size.
For example `--size 720p --seed 1` and `--size 1080p --seed 1` will produce the same image for SD15
- 🎉 feature: loading diffusers based models now supported. Example `--model https://huggingface.co/ainz/diseny-pixar --model-architecture sd15`
- 🎉 feature: qrcode controlnet!
- feature: generate word images automatically. great for use with qrcode controlnet: `imagine "flowers" --gif --size hd --control-mode qrcode --control-image "textimg='JOY' font_color=white background_color=gray" -r 10`
- feature: opendalle 1.1 added. `--model opendalle` to use it
- feature: added `--size` parameter for more intuitive sizing (e.g. 512, 256x256, 4k, uhd, FHD, VGA, etc)
- feature: detect if wrong torch version is installed and provide instructions on how to install proper version
- fix: compilation animations create normal slideshows instead of "bounces"
- fix: file globbing works in the interactive shell
- fix: fix model downloads that were broken by [library change in transformers 4.27.0](https://github.com/huggingface/transformers/commit/8f3b4a1d5bd97045541c43179efe8cd9c58adb76)
**11.0.0**
- all these changes together mean same seed/sampler will not be guaranteed to produce same image (thus the version bump)
- fix: image composition didn't work very well. Works well now but probably very slow on non-cuda platforms
Previously the `+` and `-` characters in a mask (example: `face{+0.1}`) added to the grayscale value of any masked areas. This wasn't very useful. The new behavior is that the mask will expand or contract by the number of pixel specified. The technical terms for this are dilation and erosion. This allows much greater control over the masked area.
- feature: update k-diffusion samplers. add k_dpm_adaptive and k_dpm_fast
- feature: img2img/inpainting supported on all samplers
- feature: make (blip) photo captions more descriptive
**2.3.1**
- fix: face fidelity default was broken
**2.3.0**
- feature: model weights file can be specified via `--model-weights-path` argument at the command line
- fix: set face fidelity default back to old value
- fix: handle small images without throwing exception. credit to @NiclasEriksen
- docs: add setuptools-rust as dependency for macos
**2.2.1**
- fix: init image is fully ignored if init-image-strength = 0
**2.2.0**
- feature: face enhancement fidelity is now configurable
**2.1.0**
- [improved masking accuracy from clipseg](https://github.com/timojl/clipseg/issues/8#issuecomment-1259150865)
**2.0.3**
- fix memory leak in face enhancer
- fix blurry inpainting
- fix for pillow compatibility
**2.0.0**
- 🎉 fix: inpainted areas correlate with surrounding image, even at 100% generation strength. Previously if the generation strength was high enough the generated image
would be uncorrelated to the rest of the surrounding image. It created terrible looking images.
- 🎉 feature: interactive prompt added. access by running `aimg`
- 🎉 feature: Specify advanced text based masks using boolean logic and strength modifiers. Mask descriptions must be lowercase. Keywords uppercase.
Valid symbols: `AND`, `OR`, `NOT`, `()`, and mask strength modifier `{+0.1}` where `+` can be any of `+ - * /`. Single character boolean operators also work (`|`, `&`, `!`)
- 🎉 feature: apply mask edits to original files with `mask_modify_original` (on by default)
- feature: auto-rotate images if exif data specifies to do so
- fix: mask boundaries are more accurate
- fix: accept mask images in command line
- fix: img2img algorithm was wrong and wouldn't at values close to 0 or 1
**1.6.2**
- fix: another bfloat16 fix
**1.6.1**
- fix: make sure image tensors come to the CPU as float32 so there aren't compatibility issues with non-bfloat16 cpus
**1.6.0**
- fix: *maybe* address #13 with `expected scalar type BFloat16 but found Float`
- at minimum one can specify `--precision full` now and that will probably fix the issue
- feature: tile mode can now be specified per-prompt
**1.5.3**
- fix: missing config file for describe feature
**1.5.1**
- img2img now supported with PLMS (instead of just DDIM)
- added image captioning feature `aimg describe dog.jpg` => `a brown dog sitting on grass`
- added new commandline tool `aimg` for additional image manipulation functionality
**1.4.0**
- support multiple additive targets for masking with `|` symbol. Example: "fruit|stem|fruit stem"