1faea372f9
fix: cleanup logging - remove unnecessary version checks
1 month ago
964dd4ead7
feature: integrates spandrel for upscaling
1 month ago
3a9a3974ce
fix: allow referencing local paths for sdxl model weights
...
Addresses https://github.com/brycedrennan/imaginAIry/issues/484
1 month ago
df86aa6668
feature: densepose controlnet ( #481 )
2 months ago
49f2c25b6b
feature: IP-Adapter ( #477 )
...
todo
- allow specification ip adapter weights/arch
---------
Co-authored-by: jaydrennan <jsdman1313@gmail.com>
3 months ago
9c48b749d8
feature: script for running imaginairy in the modal.com cloud
3 months ago
9cdacd454f
style: use latest ruff
3 months ago
e6a1c988c5
fix: if weights are float32 but float16 was specified, still use float16
4 months ago
601a112dc3
refactor: move download related functions to separate module ( #453 )
...
+ renames and typehints
5 months ago
502ffbdc63
feature: sdxl inpaint support ( #450 )
5 months ago
700cb457b9
feature: support loading sdxl compvis weights ( #449 )
5 months ago
907e80d1f2
feature: video interpolation ( #448 )
...
- uses rife algorithm to interpolate frames
5 months ago
bb2dd45cf2
feature: videogen improvements ( #447 )
...
- allow generation at any size
- output "bounce" animations
- choose output format: mp4, webp, or gif
- fix random seed handling
- organize some code better
5 months ago
d3106fc9e3
fix: videogen bug ( #443 )
5 months ago
ed40a12c01
version: 14.0.0
5 months ago
55e27160f5
build: vendorize refiners
...
so we can still work in conda envs
5 months ago
f84406f12c
fix: handle unexpected keys in weights better
5 months ago
5b3b04b877
build: remove pytorch lightning dependency
5 months ago
4fcfc363af
fix: always show total last
5 months ago
f50a1f5b0c
fix: interrupted generations don't prevent more generations
...
fixes #424
- pref: improve memory usage when loading SD15.
- feature: clean up CLI output more
- feature: cuda memory tracking context manager
- feature: use safetensors fp16 for sd15
5 months ago
9e3403df89
feature: clean up terminal output
...
- recording timing and memory usage of various steps
- re-use logging context for composition images
- load sdxl weights in a more VRAM efficient way
- switch to diffusers weights for default weights for sd15
5 months ago
77c4b85037
perf: improve memory usage ( #433 )
...
add warning for corrupt weights files
5 months ago
42a045e8e6
feature: support sdxl ( #431 )
...
- adds support for (SDXL)[https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0 ]
- adds sliced encoding/decoding to refiners sdxl pipeline
- doesn't support inpainting, controlnets
- monkeypatches self_attention_guidance to use sliced attention
- adds a bunch of model weight translation utilities and weightmaps
- add [opendalle 1.1](https://huggingface.co/dataautogpt3/OpenDalleV1.1 )
- change default model to opendalle
- fix: better handle special characters in path inputs on command line
**todo**
- add tests
5 months ago
7eef3bf628
feature: replaces black formatter with ruff formatter
5 months ago
a2c38b3ec0
feature: support loading diffusers folder/based models from huggingface ( #427 )
5 months ago
6ebd12abb1
refactor: move code to more intuitive places
5 months ago
616f686ed2
small changes ( #425 )
...
* docs: update todo
* refactor: small cleanup of tiling code
5 months ago
0c03612d44
feature: large images now stay well-composed thanks to tile controlnet
6 months ago
2372a71e6c
fix: adds tile/detail controlnet back in.
6 months ago
7880ee1389
feature: update midas (depth maps)
6 months ago
2144f26fa7
feature: add ability to dynamically make word images ( #417 )
6 months ago
6d39d791b1
refactor: move safety to utils
6 months ago
ad561e8833
refactor: move model_manager to utils
6 months ago
d478771cc0
refactor: move a bunch of stuff to utils
6 months ago
316114e660
docs: add docstrings
...
Wrote an openai script and custom prompt to generate them.
6 months ago
37ecd1e5e0
fix: videogen. track gpu tests
6 months ago
eae4f20ae2
ci: add type checker
...
fix some typehint issues
6 months ago
2bd6cb264b
feature: large refactor
...
- add type hints
- size parameter
- ControlNetInput => ControlInput
- simplify imagineresult
6 months ago
24f4af3482
feature: better torch installation experience
6 months ago
71d4992dca
feature: added `--size` parameter to allow using named sizes
6 months ago
0fe3733933
fix: memory management issue
...
the dtype being used as a cache key wasn't consistent. this caused the model to be loaded twice
6 months ago
e8fe8d7d6c
feature: stable diffusion video (SVD)
6 months ago
f97f6a3b4b
feature: use refiners library for generation
...
BREAKING CHANGE
- stable diffusion 1.5 + inpainting working
- self-attention guidance working. improves image generation quality
- tile-mode working
- inpainting self-attention guidance working
disable/broken features:
- sd 1.4, 2.0, 2.1
- most of the samplers
- pix2pix edit
- most of the controlnets
- memory management
- python 3.8 support
wip
6 months ago
558d3388e5
style: speed up linting and autoformatting. fix lints
8 months ago
7c2004bfcc
feature/fix: migrate to pydantic 2.3
...
- test: add schema tests/fuzzer and fixes
- fix default prompt. add tests
- fix outpaint and controlnet defaults
- fix init image strength defaults
9 months ago
c5c90df337
feature: fix debug level logging
1 year ago
e53459a50a
build: check for torch version at runtime ( fixes #329 )
1 year ago
38ac0b7f54
tests: fix tests
1 year ago
bcaa000d35
fix: model logging
1 year ago
926692ad03
tests: "prime" the controlnets
...
Trying to get things working on m1. doesn't fix everything
1 year ago
Bryce
jaydrennan