version: 15.0.0

works on my nvidia 4090 and probably not many other places. rush job. will not be providing support but will accept pull requests

.dockerignore

@@ -18,4 +18,6 @@ coverage.xml
 .mypy_cache
 .pytest_cache
 .hypothesis
-.DS_Store
+.DS_Store
+other
+outputs

.github/workflows/ci.yaml

@@ -5,47 +5,73 @@ on:
     branches:
       - master
   workflow_dispatch:
+env:
+  PIP_DISABLE_PIP_VERSION_CHECK: 1
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+  cancel-in-progress: true
+
 jobs:
   lint:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v3
-    - uses: actions/setup-python@v4
-      with:
-        python-version: 3.9
-        cache: pip
-        cache-dependency-path: requirements-dev.txt
-    - name: Install dependencies
-      run: |
-        python -m pip install --disable-pip-version-check wheel pip-tools
-        pip-sync requirements-dev.txt
-        python -m pip install --disable-pip-version-check --no-deps .
-    - name: Lint
-      run: |
-        echo "::add-matcher::.github/pylama_matcher.json"
-        pylama --options tox.ini
-  autoformat:
-    runs-on: ubuntu-latest
+      - uses: actions/checkout@v3
+      - uses: actions/setup-python@v4.5.0
+        with:
+          python-version: "3.10"
+      - name: Cache dependencies
+        uses: actions/cache@v3.2.4
+        id: cache
+        with:
+          path: ${{ env.pythonLocation }}
+          key: ${{ env.pythonLocation }}-${{ hashFiles('requirements-dev.txt') }}-lint
+      - name: Install Ruff
+        if: steps.cache.outputs.cache-hit != 'true'
+        run: grep -E 'ruff==' requirements-dev.txt | xargs pip install
+      - name: Format
+        run: |
+          echo "::add-matcher::.github/pylama_matcher.json"
+          ruff format --config tests/ruff.toml .  --check
+      - name: Lint
+        run: |
+          echo "::add-matcher::.github/pylama_matcher.json"
+          ruff check --config tests/ruff.toml .
+  test-gpu:
+    runs-on: nvidia-4090
     steps:
     - uses: actions/checkout@v3
-    - uses: actions/setup-python@v4
+    - name: Set up Python
+      uses: actions/setup-python@v4
       with:
-        python-version: 3.9
+        python-version: "3.10"
     - name: Install dependencies
       run: |
-        python -m pip install --disable-pip-version-check black==23.1.0 isort==5.12.0
-    - name: Autoformatter
+        python -m pip install uv
+        uv pip uninstall --system torch torchvision xformers triton imaginairy
+        uv pip sync --system requirements-dev.txt
+        pip install -e .
+    - name: Test with pytest
+      timeout-minutes: 30
+      env:
+        CUDA_LAUNCH_BLOCKING: 1
       run: |
-        black --diff .
-        isort --atomic --profile black --check-only .
-  test:
+        pytest --durations=10 -v -m "gputest"
+  test-non-gpu:
     runs-on: ubuntu-latest
     strategy:
+      fail-fast: false
       matrix:
-        python-version: ["3.8", "3.10"]
+        python-version:
+          - "3.10"
+          - "3.11"
+# torch for python 3.12 is not available yet
+# https://github.com/pytorch/pytorch/issues/110436
+#          - "3.12"
+        os: ["ubuntu-latest"]
     steps:
     - uses: actions/checkout@v3
-    - name: Set up Python ${{ matrix.python-version }}
+    - name: Set up Python
       uses: actions/setup-python@v4
       with:
         python-version: ${{ matrix.python-version }}
@@ -53,22 +79,156 @@ jobs:
         cache-dependency-path: requirements-dev.txt
     - name: Install dependencies
       run: |
-        python -m pip install --disable-pip-version-check -r requirements-dev.txt
-        python -m pip install --disable-pip-version-check .
-    - name: Get current date
-      id: date
-      run: echo "::set-output name=curmonth::$(date +'%Y-%m')"
-    - name: Cache Model Files
-      id: cache-model-files
-      uses: actions/cache@v3
-      with:
-        path: |
-          ~/.cache/huggingface
-          ~/.cache/clip
-          ~/.cache/imaginairy
-          ~/.cache/torch
-        key: ${{ steps.date.outputs.curmonth }}-b
+        python -m pip install uv 
+        uv pip sync --system requirements-dev.txt
+        pip install -e .
     - name: Test with pytest
-      timeout-minutes: 20
+      timeout-minutes: 30
+      env:
+        CUDA_LAUNCH_BLOCKING: 1
       run: |
-        pytest --durations=50 -v
+        pytest --durations=10 -v -m "not gputest"
+  type-check:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v3
+    - uses: actions/setup-python@v4.5.0
+      with:
+        python-version: "3.10"
+        cache: pip
+        cache-dependency-path: requirements-dev.txt
+    - name: Install dependencies
+      run: |
+        python -m pip install -r requirements-dev.txt . --upgrade
+    - name: Run mypy
+      run: |
+        make type-check
+  build-wheel:
+    name: Build Wheel
+    runs-on: ubuntu-latest
+    outputs:
+      wheel_name: ${{ steps.set_wheel_name.outputs.wheel_name }}
+    steps:
+      - uses: actions/checkout@v3
+      - uses: actions/setup-python@v4.5.0
+        with:
+          python-version: "3.10"
+      - name: Install dependencies
+        run: python -m pip install wheel
+      - name: Build package
+        run: python setup.py bdist_wheel
+      - name: Set wheel filename
+        id: set_wheel_name
+        run: echo "wheel_name=$(ls dist/*.whl)" >> "$GITHUB_OUTPUT"
+      - uses: actions/upload-artifact@v3
+        with:
+          name: wheels
+          path: dist/*.whl
+  smoke-test-wheel:
+    needs: build-wheel
+    name:  Smoketest (Python ${{ matrix.python-version }}, ${{ matrix.os }})
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        python-version: ["3.10", "3.11"]
+        os: ["ubuntu-latest", "windows-latest", "m2-16gb"]
+    steps:
+    - uses: actions/checkout@v3
+    - uses: actions/setup-python@v4.5.0
+      with:
+        python-version: ${{ matrix.python-version }}
+    - uses: actions/download-artifact@v3
+      with:
+        name: wheels
+        path: dist
+    - name: Install built wheel
+      env:
+        WHEEL_FILENAME: ${{ needs.build-wheel.outputs.wheel_name }}
+      run: |
+        python -m pip install uv
+        uv pip install --system ${{ needs.build-wheel.outputs.wheel_name }}
+    - name: Generate an image
+      run: |
+        imagine fruit --steps 3 --size 128 --seed 1
+    - uses: actions/upload-artifact@v3
+      with:
+        name: images
+        path: outputs/generated/*.jpg
+  smoke-test-wheel-conda:
+    needs: build-wheel
+    name: Smoketest (Conda, Python ${{ matrix.python-version }}, ${{ matrix.os }})
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        python-version: [ "3.10", "3.11" ]
+        os: [ "ubuntu-latest", "windows-latest", "m2-16gb" ]
+    steps:
+      - uses: actions/checkout@v3
+      - uses: conda-incubator/setup-miniconda@v3
+        with:
+          auto-update-conda: true
+          python-version: ${{ matrix.python-version }}
+          activate-environment: test-env
+          create-env-file: true
+      - uses: actions/download-artifact@v3
+        with:
+          name: wheels
+          path: dist
+      - name: Install built wheel
+        shell: bash -l {0}
+        env:
+          WHEEL_FILENAME: ${{ steps.set-wheel-name.outputs.WHEEL_FILENAME }}
+        run: |
+          conda activate test-env
+          python -m pip install uv
+          uv pip install ${{ needs.build-wheel.outputs.wheel_name }}
+      - name: Generate an image
+        shell: bash -l {0}
+        run: |
+          conda activate test-env
+          imagine fruit --steps 3 --size 128 --seed 1
+      - uses: actions/upload-artifact@v3
+        with:
+          name: images
+          path: outputs/generated/*.jpg
+
+  # build-docs:
+  #   runs-on: ubuntu-latest
+  #   steps:
+  #     - uses: actions/checkout@v4
+  #     - uses: actions/setup-python@v4
+  #       with:
+  #         python-version: "3.10"
+  #     - run: echo "cache_id=$(date --utc '+%V')" >> $GITHUB_ENV
+  #     - uses: actions/cache@v3
+  #       with:
+  #         key: mkdocs-material-${{ env.cache_id }}
+  #         path: .cache
+  #         restore-keys: |
+  #           mkdocs-material-
+  #     - run: python -m pip install -r requirements-dev.in . --upgrade
+  #     - run: mkdocs build --strict
+
+  publish-docs:
+    if: github.ref == 'refs/heads/master'
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - name: Configure Git Credentials
+        run: |
+          git config user.name github-actions[bot]
+          git config user.email 41898282+github-actions[bot]@users.noreply.github.com
+      - uses: actions/setup-python@v4
+        with:
+          python-version: "3.10"
+      - run: echo "cache_id=$(date --utc '+%V')" >> $GITHUB_ENV
+      - uses: actions/cache@v3
+        with:
+          key: mkdocs-material-${{ env.cache_id }}
+          path: .cache
+          restore-keys: |
+            mkdocs-material-
+      - run: python -m pip install -r requirements-dev.in . --upgrade
+      - run: mkdocs gh-deploy --force

.gitignore

@@ -30,3 +30,6 @@ tests/vastai_cli.py
 **/.eggs
 /img_size_memory_usage.csv
 /tests/test_cluster_output/
+/.env
+*.ipynb
+.ipynb_checkpoints

Makefile

@@ -1,5 +1,5 @@
 SHELL := /bin/bash
-python_version = 3.10.10
+python_version = 3.10.13
 venv_prefix = imaginairy
 venv_name = $(venv_prefix)-$(python_version)
 pyenv_instructions=https://github.com/pyenv/pyenv#installation
@@ -9,49 +9,63 @@ pyenv_virt_instructions=https://github.com/pyenv/pyenv-virtualenv#pyenv-virtuale
 init: require_pyenv  ## Setup a dev environment for local development.
 	@pyenv install $(python_version) -s
 	@echo -e "\033[0;32m ✔️  🐍 $(python_version) installed \033[0m"
-	@if ! [ -d "$$(pyenv root)/versions/$(venv_name)" ]; then\
-		pyenv virtualenv $(python_version) $(venv_name);\
-	fi;
+	@if ! [ -d "$$(pyenv root)/versions/$(venv_name)" ]; then \
+		pyenv virtualenv $(python_version) $(venv_name); \
+	fi
 	@pyenv local $(venv_name)
 	@echo -e "\033[0;32m ✔️  🐍 $(venv_name) virtualenv activated \033[0m"
-	pip install --upgrade pip pip-tools
-	pip-sync requirements-dev.txt
-	pip install -e . --no-deps
-	# the compiled requirements don't included OS specific subdependencies so we trigger those this way
-	pip install `pip freeze | grep "^torch=="`
-	@echo -e "\nEnvironment setup! ✨ 🍰 ✨ 🐍 \n\nCopy this path to tell PyCharm where your virtualenv is. You may have to click the refresh button in the pycharm file explorer.\n"
-	@echo -e "\033[0;32m"
-	@pyenv which python
-	@echo -e "\n\033[0m"
-	@echo -e "The following commands are available to run in the Makefile\n"
+	@export VIRTUAL_ENV=$$(pyenv prefix); \
+	if command -v uv >/dev/null 2>&1; then \
+		uv pip install --upgrade uv; \
+	else \
+		pip install --upgrade pip uv; \
+	fi; \
+	uv pip sync requirements-dev.txt; \
+	uv pip install -e .
+	@echo -e "\nEnvironment setup! ✨ 🍰 ✨ 🐍 \n\nCopy this path to tell PyCharm where your virtualenv is. You may have to click the refresh button in the PyCharm file explorer.\n"
+	@echo -e "\033[0;32m$$(pyenv which python)\033[0m\n"
+	@echo -e "The following commands are available to run in the Makefile:\n"
 	@make -s help
 
+
 af: autoformat  ## Alias for `autoformat`
 autoformat:  ## Run the autoformatter.
-	@pycln . --all --quiet --extend-exclude __init__\.py
-	@# ERA,T201
-	@-ruff --extend-ignore ANN,ARG001,C90,DTZ,D100,D101,D102,D103,D202,D203,D212,D415,E501,RET504,S101,UP006,UP007 --extend-select C,D400,I,W --unfixable T,ERA --fix-only .
-	@black .
-	@isort --atomic --profile black .
+	@-ruff check --config tests/ruff.toml . --fix-only
+	@ruff format --config tests/ruff.toml .
 
 test:  ## Run the tests.
 	@pytest
 	@echo -e "The tests pass! ✨ 🍰 ✨"
 
+test-fast:  ## Run the fast tests.
+	@pytest -m "not gputest"
+	@echo -e "The non-gpu tests pass! ✨ 🍰 ✨"
+
 lint:  ## Run the code linter.
-	@pylama
+	@ruff check --config tests/ruff.toml .
 	@echo -e "No linting errors - well done! ✨ 🍰 ✨"
 
+type-check: ## Run the type checker.
+	@mypy --config-file tox.ini .
+
+check-fast:  ## Run autoformatter, linter, typechecker, and fast tests
+	@make autoformat
+	@make lint
+	@make type-check
+	@make test-fast
+
+build-pkg:  ## Build the package
+	python setup.py sdist bdist_wheel
+	python setup.py bdist_wheel --plat-name=win-amd64
+
 deploy:  ## Deploy the package to pypi.org
 	pip install twine wheel
 	-git tag $$(python setup.py -V)
 	git push --tags
 	rm -rf dist
-	python setup.py bdist_wheel
-	python setup.py bdist_wheel --plat-name=win-amd64
+	make build-pkg
 	#python setup.py sdist
-	@echo 'pypi.org Username: '
-	@read username && twine upload --verbose dist/* -u $$username;
+	@twine upload --verbose dist/* -u __token__;
 	rm -rf build
 	rm -rf dist
 	@echo "Deploy successful! ✨ 🍰 ✨"
@@ -72,12 +86,24 @@ require_pyenv:
 	else\
 	  echo -e "\033[0;32m ✔️  pyenv installed\033[0m";\
 	fi
-	@if ! [[ "$$(pyenv virtualenv --version)" == *"pyenv-virtualenv"* ]]; then\
-	  echo -e '\n\033[0;31m ❌ pyenv virtualenv is not installed.  Follow instructions here: $(pyenv_virt_instructions) \n\033[0m';\
-	  exit 1;\
-	else\
-	  echo -e "\033[0;32m ✔️  pyenv-virtualenv installed\033[0m";\
-	fi
+
+.PHONY: docs
+
+docs:
+	mkdocs serve
+
+update-stablestudio:
+	@echo "Updating stablestudio"
+	cd ../imaginAIry-StableStudio && \
+	yarn build && \
+	yarn build:production
+	rm -rf imaginairy/http/stablestudio/dist
+	cp -R ../imaginAIry-StableStudio/packages/stablestudio-ui/dist imaginairy/http/stablestudio/dist
+	rm -rf imaginairy/http/stablestudio/dist/examples
+	rm -rf imaginairy/http/stablestudio/dist/media
+	rm -rf imaginairy/http/stablestudio/dist/presets
+	cp ../imaginAIry-StableStudio/LICENSE imaginairy/http/stablestudio/dist/LICENSE
+	@echo "Updated stablestudio"
 
 vendor_openai_clip:
 	mkdir -p ./downloads
@@ -171,13 +197,39 @@ vendorize_controlnet_annotators:
 
 
 
-vendorize_surface_normal_uncertainty:
-	make download_repo REPO=git@github.com:baegwangbin/surface_normal_uncertainty.git PKG=surface_normal_uncertainty COMMIT=fe2b9f1e8a4cac1c73475b023f6454bd23827a48
-	mkdir -p ./imaginairy/vendored/surface_normal_uncertainty
-	rm -rf ./imaginairy/vendored/surface_normal_uncertainty/*
-	cp -R ./downloads/surface_normal_uncertainty/* ./imaginairy/vendored/surface_normal_uncertainty/
+vendorize_normal_map:
+	make download_repo REPO=git@github.com:brycedrennan/imaginairy-normal-map.git PKG=imaginairy_normal_map COMMIT=6b3b1692cbdc21d55c84a01e0b7875df030b6d79
+	mkdir -p ./imaginairy/vendored/imaginairy_normal_map
+	rm -rf ./imaginairy/vendored/imaginairy_normal_map/*
+	cp -R ./downloads/imaginairy_normal_map/imaginairy_normal_map/* ./imaginairy/vendored/imaginairy_normal_map/
+	make af
 
 
+vendorize_refiners:
+	export REPO=git@github.com:finegrain-ai/refiners.git PKG=refiners COMMIT=91aea9b7ff63ddf93f99e2ce6a4452bd658b1948 && \
+	make download_repo REPO=$$REPO PKG=$$PKG COMMIT=$$COMMIT && \
+	mkdir -p ./imaginairy/vendored/$$PKG && \
+	rm -rf ./imaginairy/vendored/$$PKG/* && \
+	cp -R ./downloads/refiners/src/refiners/* ./imaginairy/vendored/$$PKG/ && \
+	cp ./downloads/refiners/LICENSE ./imaginairy/vendored/$$PKG/ && \
+	rm -rf ./imaginairy/vendored/$$PKG/training_utils && \
+	echo "vendored from $$REPO @ $$COMMIT" | tee ./imaginairy/vendored/$$PKG/readme.txt
+	find ./imaginairy/vendored/refiners/ -type f -name "*.py" -exec sed -i '' 's/from refiners/from imaginairy.vendored.refiners/g' {} + &&\
+    find ./imaginairy/vendored/refiners/ -type f -name "*.py" -exec sed -i '' 's/import refiners/import imaginairy.vendored.refiners/g' {} + &&\
+	make af
+
+vendorize_facexlib:
+	export REPO=git@github.com:xinntao/facexlib.git PKG=facexlib COMMIT=260620ae93990a300f4b16448df9bb459f1caba9 && \
+	make download_repo REPO=$$REPO PKG=$$PKG COMMIT=$$COMMIT && \
+	mkdir -p ./imaginairy/vendored/$$PKG && \
+	rm -rf ./imaginairy/vendored/$$PKG/* && \
+	cp -R ./downloads/$$PKG/facexlib/* ./imaginairy/vendored/$$PKG/ && \
+	rm -rf ./imaginairy/vendored/$$PKG/weights && \
+	cp ./downloads/$$PKG/LICENSE ./imaginairy/vendored/$$PKG/ && \
+	echo "vendored from $$REPO @ $$COMMIT" | tee ./imaginairy/vendored/$$PKG/readme.txt
+	find ./imaginairy/vendored/facexlib/ -type f -name "*.py" -exec sed -i '' 's/from facexlib/from imaginairy.vendored.facexlib/g' {} + &&\
+	sed -i '' '/from \.version import __gitsha__, __version__/d' ./imaginairy/vendored/facexlib/__init__.py
+	make af
 
 vendorize:  ## vendorize a github repo.  `make vendorize REPO=git@github.com:openai/CLIP.git PKG=clip`
 	mkdir -p ./downloads

README.md

@@ -4,31 +4,134 @@
 [![Downloads](https://pepy.tech/badge/imaginairy)](https://pepy.tech/project/imaginairy)
 [![image](https://img.shields.io/pypi/v/imaginairy.svg)](https://pypi.org/project/imaginairy/)
 [![image](https://img.shields.io/badge/license-MIT-green)](https://github.com/brycedrennan/imaginAIry/blob/master/LICENSE/)
-[![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/ambv/black)
-[![Python Checks](https://github.com/brycedrennan/imaginAIry/actions/workflows/ci.yaml/badge.svg)](https://github.com/brycedrennan/imaginAIry/actions/workflows/ci.yaml)
+[![Discord](https://flat.badgen.net/discord/members/FdD7ut3YjW)](https://discord.gg/FdD7ut3YjW)
 
-AI imagined images. Pythonic generation of stable diffusion images.
+AI imagined images. Pythonic generation of stable diffusion images **and videos** *!.
 
-"just works" on Linux and macOS(M1) (and maybe windows?).
+"just works" on Linux and macOS(M1) (and sometimes windows).
 
 
 ```bash
 # on macOS, make sure rust is installed first
+# be sure to use Python 3.10, Python 3.11 is not supported at the moment
 >> pip install imaginairy
 >> imagine "a scenic landscape" "a photo of a dog" "photo of a fruit bowl" "portrait photo of a freckled woman" "a bluejay"
-# Make an animation showing the generation process
->> imagine --gif "a flower"
+# Make an AI video
+>> aimg videogen --start-image rocket.png
 ```
+## Stable Video Diffusion
 <p float="left">
-<img src="assets/000019_786355545_PLMS50_PS7.5_a_scenic_landscape.jpg" height="256">
-<img src="assets/000032_337692011_PLMS40_PS7.5_a_photo_of_a_dog.jpg" height="256">
-<img src="assets/000056_293284644_PLMS40_PS7.5_photo_of_a_bowl_of_fruit.jpg" height="256">
-<img src="assets/000078_260972468_PLMS40_PS7.5_portrait_photo_of_a_freckled_woman.jpg" height="256">
-<img src="assets/013986_1_kdpmpp2m59_PS7.5_a_bluejay_[generated].jpg" height="256">
-<img src="assets/009719_942389026_kdpmpp2m15_PS7.5_a_flower.gif" height="256">
+<img src="docs/assets/svd-rocket.gif" height="190">
+<img src="docs/assets/svd-athens.gif" height="190">
+<img src="docs/assets/svd-pearl-girl.gif" height="190">
+<img src="docs/assets/svd-starry-night.gif" height="190">
+<img src="docs/assets/svd-dog.gif" height="190">
+<img src="docs/assets/svd-xpbliss.gif" height="190">
 </p>
 
+### Rushed release of Stable Diffusion Video!
+Works with Nvidia GPUs.  Does not work on Mac or CPU.
+
+On Windows you'll need to install torch 2.0 first via https://pytorch.org/get-started/locally/
+```text
+Usage: aimg videogen [OPTIONS]
+
+  AI generate a video from an image
+
+  Example:
+
+      aimg videogen --start-image assets/rocket-wide.png
+
+Options:
+  --start-image TEXT       Input path for image file.
+  --num-frames INTEGER     Number of frames.
+  --num-steps INTEGER      Number of steps.
+  --model TEXT             Model to use. One of: svd, svd_xt, svd_image_decoder, svd_xt_image_decoder
+  --fps INTEGER            FPS for the AI to target when generating video
+  --output-fps INTEGER     FPS for the output video
+  --motion-amount INTEGER  How much motion to generate. value between 0 and 255.
+  -r, --repeats INTEGER    How many times to repeat the renders.   [default: 1]
+  --cond-aug FLOAT         Conditional augmentation.
+  --seed INTEGER           Seed for random number generator.
+  --decoding_t INTEGER     Number of frames decoded at a time.
+  --output_folder TEXT     Output folder.
+  --help                   Show this message and exit.
+```
+
+### Images
+<p float="left">
+<img src="docs/assets/026882_1_ddim50_PS7.5_a_scenic_landscape_[generated].jpg" height="256">
+<img src="docs/assets/026884_1_ddim50_PS7.5_photo_of_a_dog_[generated].jpg" height="256">
+<img src="docs/assets/026890_1_ddim50_PS7.5_photo_of_a_bowl_of_fruit._still_life_[generated].jpg" height="256">
+<img src="docs/assets/026885_1_ddim50_PS7.5_girl_with_a_pearl_earring_[generated].jpg" height="256">
+<img src="docs/assets/026891_1_ddim50_PS7.5_close-up_photo_of_a_bluejay_[generated].jpg" height="256">
+<img src="docs/assets/026893_1_ddim50_PS7.5_macro_photo_of_a_flower_[generated].jpg" height="256">
+</p>
+
+### Whats New
+[See full Changelog here](./docs/changelog.md)
+
+
+**14.3.0**
+- feature: integrates [spandrel](https://github.com/chaiNNer-org/spandrel) for upscaling 
+- fix: allow loading sdxl models from local paths. 
+
+**14.2.0**
+- 🎉 feature: add image prompt support via `--image-prompt` and `--image-prompt-strength`
+
+**14.1.1**
+- tests: add installation tests for windows, mac, and conda
+- fix: dependency issues
+
+**14.1.0**
+- 🎉 feature: make video generation smooth by adding frame interpolation
+- feature: SDXL weights in the compvis format can now be used
+- feature: allow video generation at any size specified by user
+- feature: video generations output in "bounce" format
+- feature: choose video output format: mp4, webp, or gif
+- feature: fix random seed handling in video generation
+- docs: auto-publish docs on push to master
+- build: remove imageio dependency
+- build: vendorize facexlib so we don't install its unneeded dependencies
+
+
+**14.0.4**
+- docs: add a documentation website at https://brycedrennan.github.io/imaginAIry/
+- build: remove fairscale dependency
+- fix: video generation was broken
+
+**14.0.3**
+- fix: several critical bugs with package
+- tests: add a wheel smoketest to detect these issues in the future
+
+**14.0.0**
+- 🎉 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!
+
+
+### Run API server and StableStudio web interface (alpha)
+Generate images via API or web interface.  Much smaller featureset compared to the command line tool.
+```bash
+>> aimg server
+```
+Visit http://localhost:8000/ and http://localhost:8000/docs
+
+<img src="https://github.com/Stability-AI/StableStudio/blob/a65d4877ad7d309627808a169818f1add8c278ae/misc/GenerateScreenshot.png?raw=true" width="512">
+
 ### Image Structure Control [by ControlNet](https://github.com/lllyasviel/ControlNet)
+#### (Not supported for SDXL yet)
 Generate images guided by body poses, depth maps, canny edges, hed boundaries, or normal maps.
 
 **Openpose Control**
@@ -38,60 +141,60 @@ imagine --control-image assets/indiana.jpg  --control-mode openpose --caption-te
 ```
 
 <p float="left">
-    <img src="assets/indiana.jpg" height="256">
-    <img src="assets/indiana-pose.jpg" height="256">
-    <img src="assets/indiana-pose-polar-bear.jpg" height="256">
+    <img src="docs/assets/indiana.jpg" height="256">
+    <img src="docs/assets/indiana-pose.jpg" height="256">
+    <img src="docs/assets/indiana-pose-polar-bear.jpg" height="256">
 </p>
 
-**Canny Edge Control**
+#### Canny Edge Control
 
 ```bash
 imagine --control-image assets/lena.png  --control-mode canny "photo of a woman with a hat looking at the camera"
 ```
 
 <p float="left">
-    <img src="assets/lena.png" height="256">
-    <img src="assets/lena-canny.jpg" height="256">
-    <img src="assets/lena-canny-generated.jpg" height="256">
+    <img src="docs/assets/lena.png" height="256">
+    <img src="docs/assets/lena-canny.jpg" height="256">
+    <img src="docs/assets/lena-canny-generated.jpg" height="256">
 </p>
 
-**HED Boundary Control**
+#### HED Boundary Control
 
 ```bash
 imagine --control-image dog.jpg  --control-mode hed  "photo of a dalmation"
 ```
 
 <p float="left">
-    <img src="assets/000032_337692011_PLMS40_PS7.5_a_photo_of_a_dog.jpg" height="256">
-    <img src="assets/dog-hed-boundary.jpg" height="256">
-    <img src="assets/dog-hed-boundary-dalmation.jpg" height="256">
+    <img src="docs/assets/000032_337692011_PLMS40_PS7.5_a_photo_of_a_dog.jpg" height="256">
+    <img src="docs/assets/dog-hed-boundary.jpg" height="256">
+    <img src="docs/assets/dog-hed-boundary-dalmation.jpg" height="256">
 </p>
 
-**Depth Map Control**
+#### Depth Map Control
 
 ```bash
 imagine --control-image fancy-living.jpg  --control-mode depth  "a modern living room"
 ```
 
 <p float="left">
-    <img src="assets/fancy-living.jpg" height="256">
-    <img src="assets/fancy-living-depth.jpg" height="256">
-    <img src="assets/fancy-living-depth-generated.jpg" height="256">
+    <img src="docs/assets/fancy-living.jpg" height="256">
+    <img src="docs/assets/fancy-living-depth.jpg" height="256">
+    <img src="docs/assets/fancy-living-depth-generated.jpg" height="256">
 </p>
 
-**Normal Map Control**
+#### Normal Map Control
 
 ```bash
 imagine --control-image bird.jpg  --control-mode normal  "a bird"
 ```
 
 <p float="left">
-    <img src="assets/013986_1_kdpmpp2m59_PS7.5_a_bluejay_[generated].jpg" height="256">
-    <img src="assets/bird-normal.jpg" height="256">
-    <img src="assets/bird-normal-generated.jpg" height="256">
+    <img src="docs/assets/013986_1_kdpmpp2m59_PS7.5_a_bluejay_[generated].jpg" height="256">
+    <img src="docs/assets/bird-normal.jpg" height="256">
+    <img src="docs/assets/bird-normal-generated.jpg" height="256">
 </p>
 
-**Image Shuffle Control**
+#### Image Shuffle Control
 
 Generates the image based on elements of the control image. Kind of similar to style transfer.
 ```bash
@@ -99,12 +202,12 @@ imagine --control-image pearl-girl.jpg  --control-mode shuffle  "a clown"
 ```
 The middle image is the "shuffled" input image
 <p float="left">
-    <img src="assets/girl_with_a_pearl_earring.jpg" height="256">
-    <img src="assets/pearl_shuffle_019331_1_kdpmpp2m15_PS7.5_img2img-0.0_a_clown.jpg" height="256">
-    <img src="assets/pearl_shuffle_clown_019331_1_kdpmpp2m15_PS7.5_img2img-0.0_a_clown.jpg" height="256">
+    <img src="docs/assets/girl_with_a_pearl_earring.jpg" height="256">
+    <img src="docs/assets/pearl_shuffle_019331_1_kdpmpp2m15_PS7.5_img2img-0.0_a_clown.jpg" height="256">
+    <img src="docs/assets/pearl_shuffle_clown_019331_1_kdpmpp2m15_PS7.5_img2img-0.0_a_clown.jpg" height="256">
 </p>
 
-**Edit Instructions Control**
+#### Editing Instructions Control
 
 Similar to instructPix2Pix (below) but works with any SD 1.5 based model.
 ```bash
@@ -112,36 +215,52 @@ imagine --control-image pearl-girl.jpg  --control-mode edit --init-image-strengt
 ```
 
 <p float="left">
-    <img src="assets/girl_with_a_pearl_earring.jpg" height="256">
-    <img src="assets/pearl_anime_019537_521829407_kdpmpp2m30_PS9.0_img2img-0.01_make_it_anime.jpg" height="256">
-    <img src="assets/pearl_beach_019561_862735879_kdpmpp2m30_PS7.0_img2img-0.01_make_it_at_the_beach.jpg" height="256">
+    <img src="docs/assets/girl_with_a_pearl_earring.jpg" height="256">
+    <img src="docs/assets/pearl_anime_019537_521829407_kdpmpp2m30_PS9.0_img2img-0.01_make_it_anime.jpg" height="256">
+    <img src="docs/assets/pearl_beach_019561_862735879_kdpmpp2m30_PS7.0_img2img-0.01_make_it_at_the_beach.jpg" height="256">
 </p>
 
-**Add Details Control (upscaling/super-resolution)**
+#### Add Details Control (upscaling/super-resolution)
 
 Replaces existing details in an image. Good to use with --init-image-strength 0.2
 ```bash
-imagine --control-image "assets/wishbone.jpg" --control-mode tile "sharp focus, high-resolution" --init-image-strength 0.2 --steps 30 -w 2048 -h 2048 
+imagine --control-image "assets/wishbone.jpg" --control-mode details "sharp focus, high-resolution" --init-image-strength 0.2 --steps 30 -w 2048 -h 2048 
 ```
 
 <p float="left">
-    <img src="assets/wishbone_headshot_badscale.jpg" height="256">
-    <img src="assets/wishbone_headshot_details.jpg" height="256">
+    <img src="docs/assets/wishbone_headshot_badscale.jpg" height="256">
+    <img src="docs/assets/wishbone_headshot_details.jpg" height="256">
 </p>
 
 
+### Image (re)Colorization (using brightness control)
+Colorize black and white images or re-color existing images.
+
+The generated colors will be applied back to the original image. You can either provide a caption or 
+allow the tool to generate one for you.
+
+```bash
+aimg colorize pearl-girl.jpg --caption "photo of a woman"
+```
+<p float="left">
+    <img src="docs/assets/girl_with_a_pearl_earring.jpg" height="256">
+    <img src="docs/assets/pearl-gray.jpg" height="256">
+    <img src="docs/assets/pearl-recolor-a.jpg" height="256">
+</p>
+
 ###  Instruction based image edits [by InstructPix2Pix](https://github.com/timothybrooks/instruct-pix2pix)
+#### (Broken as of 14.0.0)
 Just tell imaginairy how to edit the image and it will do it for you!
 <p float="left">
-<img src="assets/scenic_landscape_winter.jpg" height="256">
-<img src="assets/dog_red.jpg" height="256">
-<img src="assets/bowl_of_fruit_strawberries.jpg" height="256">
-<img src="assets/freckled_woman_cyborg.jpg" height="256">
-<img src="assets/014214_51293814_kdpmpp2m30_PS10.0_img2img-1.0_make_the_bird_wear_a_cowboy_hat_[generated].jpg" height="256">
-<img src="assets/flower-make-the-flower-out-of-paper-origami.gif" height="256">
-<img src="assets/girl-pearl-clown-compare.gif" height="256">
-<img src="assets/mona-lisa-headshot-anim.gif" height="256">
-<img src="assets/make-it-night-time.gif" height="256">
+<img src="docs/assets/scenic_landscape_winter.jpg" height="256">
+<img src="docs/assets/dog_red.jpg" height="256">
+<img src="docs/assets/bowl_of_fruit_strawberries.jpg" height="256">
+<img src="docs/assets/freckled_woman_cyborg.jpg" height="256">
+<img src="docs/assets/014214_51293814_kdpmpp2m30_PS10.0_img2img-1.0_make_the_bird_wear_a_cowboy_hat_[generated].jpg" height="256">
+<img src="docs/assets/flower-make-the-flower-out-of-paper-origami.gif" height="256">
+<img src="docs/assets/girl-pearl-clown-compare.gif" height="256">
+<img src="docs/assets/mona-lisa-headshot-anim.gif" height="256">
+<img src="docs/assets/make-it-night-time.gif" height="256">
 </p>
 
 <details>
@@ -174,12 +293,12 @@ Want just quickly have some fun? Try `edit-demo` to apply some pre-defined edits
 >> aimg edit-demo pearl_girl.jpg
 ```
 <p float="left">
-<img src="assets/girl_with_a_pearl_earring_suprise.gif" height="256">
-<img src="assets/mona-lisa-suprise.gif" height="256">
-<img src="assets/luke-suprise.gif" height="256">
-<img src="assets/spock-suprise.gif" height="256">
-<img src="assets/gg-bridge-suprise.gif" height="256">
-<img src="assets/shire-suprise.gif" height="256">
+<img src="docs/assets/girl_with_a_pearl_earring_suprise.gif" height="256">
+<img src="docs/assets/mona-lisa-suprise.gif" height="256">
+<img src="docs/assets/luke-suprise.gif" height="256">
+<img src="docs/assets/spock-suprise.gif" height="256">
+<img src="docs/assets/gg-bridge-suprise.gif" height="256">
+<img src="docs/assets/shire-suprise.gif" height="256">
 </p>
 
 
@@ -207,11 +326,11 @@ When writing strength modifiers keep in mind that pixel values are between 0 and
     --fix-faces \
     "a modern female president" "a female robot" "a female doctor" "a female firefighter"
 ```
-<img src="assets/mask_examples/pearl000.jpg" height="200">➡️ 
-<img src="assets/mask_examples/pearl_pres.png" height="200">
-<img src="assets/mask_examples/pearl_robot.png" height="200">
-<img src="assets/mask_examples/pearl_doctor.png" height="200">
-<img src="assets/mask_examples/pearl_firefighter.png" height="200">
+<img src="docs/assets/mask_examples/pearl000.jpg" height="200">➡️ 
+<img src="docs/assets/mask_examples/pearl_pres.png" height="200">
+<img src="docs/assets/mask_examples/pearl_robot.png" height="200">
+<img src="docs/assets/mask_examples/pearl_doctor.png" height="200">
+<img src="docs/assets/mask_examples/pearl_firefighter.png" height="200">
 
 ```bash
 >> imagine \
@@ -222,11 +341,11 @@ When writing strength modifiers keep in mind that pixel values are between 0 and
     --init-image-strength .1 \
     "a bowl of kittens" "a bowl of gold coins" "a bowl of popcorn" "a bowl of spaghetti"
 ```
-<img src="assets/000056_293284644_PLMS40_PS7.5_photo_of_a_bowl_of_fruit.jpg" height="200">➡️ 
-<img src="assets/mask_examples/bowl004.jpg" height="200">
-<img src="assets/mask_examples/bowl001.jpg" height="200">
-<img src="assets/mask_examples/bowl002.jpg" height="200">
-<img src="assets/mask_examples/bowl003.jpg" height="200">
+<img src="docs/assets/000056_293284644_PLMS40_PS7.5_photo_of_a_bowl_of_fruit.jpg" height="200">➡️ 
+<img src="docs/assets/mask_examples/bowl004.jpg" height="200">
+<img src="docs/assets/mask_examples/bowl001.jpg" height="200">
+<img src="docs/assets/mask_examples/bowl002.jpg" height="200">
+<img src="docs/assets/mask_examples/bowl003.jpg" height="200">
 
 
 ### Face Enhancement [by CodeFormer](https://github.com/sczhou/CodeFormer)
@@ -238,41 +357,64 @@ When writing strength modifiers keep in mind that pixel values are between 0 and
 <img src="https://github.com/brycedrennan/imaginAIry/raw/master/assets/000178_1_PLMS40_PS7.5_a_couple_smiling_fixed.png" height="256"> 
 
 
-### Upscaling [by RealESRGAN](https://github.com/xinntao/Real-ESRGAN)
-```bash
->> imagine "colorful smoke" --steps 40 --upscale
-# upscale an existing image
->> aimg upscale my-image.jpg
+## Image Upscaling
+Upscale images easily.
+
+=== "CLI"
+    ```bash
+    aimg upscale assets/000206_856637805_PLMS40_PS7.5_colorful_smoke.jpg --upscale-model real-hat
+    ```
+
+=== "Python"
+    ```py
+    from imaginairy.api.upscale import upscale
+
+    img = upscale(img="assets/000206_856637805_PLMS40_PS7.5_colorful_smoke.jpg")
+    img.save("colorful_smoke.upscaled.jpg")
+
+    ```
+<img src="docs/assets/000206_856637805_PLMS40_PS7.5_colorful_smoke.jpg" width="25%" height="auto"> ➡️ 
+<img src="docs/assets/000206_856637805_PLMS40_PS7.5_colorful_smoke_upscaled.jpg" width="50%" height="auto">
+
+Upscaling uses [Spandrel](https://github.com/chaiNNer-org/spandrel) to make it easy to use different upscaling models.
+You can view different integrated models by running `aimg upscale --list-models`, and then use it with `--upscale-model <model-name>`.
+Also accepts url's if you want to upscale an image with a different model. Control the new file format/location with --format.
+
+```python
+from imaginairy.enhancers.upscale_realesrgan import upscale_image
+from PIL import Image
+img = Image.open("my-image.jpg")
+big_img = upscale_image(i)
 ```
-<img src="https://github.com/brycedrennan/imaginAIry/raw/master/assets/000206_856637805_PLMS40_PS7.5_colorful_smoke.jpg" height="128"> ➡️ 
-<img src="https://github.com/brycedrennan/imaginAIry/raw/master/assets/000206_856637805_PLMS40_PS7.5_colorful_smoke_upscaled.jpg" height="256"> 
+
+
 
 ### Tiled Images
 ```bash
 >> imagine  "gold coins" "a lush forest" "piles of old books" leaves --tile
 ```
 
-<img src="assets/000066_801493266_PLMS40_PS7.5_gold_coins.jpg" height="128"><img src="assets/000066_801493266_PLMS40_PS7.5_gold_coins.jpg" height="128"><img src="assets/000066_801493266_PLMS40_PS7.5_gold_coins.jpg" height="128">
-<img src="assets/000118_597948545_PLMS40_PS7.5_a_lush_forest.jpg" height="128"><img src="assets/000118_597948545_PLMS40_PS7.5_a_lush_forest.jpg" height="128"><img src="assets/000118_597948545_PLMS40_PS7.5_a_lush_forest.jpg" height="128">
+<img src="docs/assets/000066_801493266_PLMS40_PS7.5_gold_coins.jpg" height="128"><img src="docs/assets/000066_801493266_PLMS40_PS7.5_gold_coins.jpg" height="128"><img src="docs/assets/000066_801493266_PLMS40_PS7.5_gold_coins.jpg" height="128">
+<img src="docs/assets/000118_597948545_PLMS40_PS7.5_a_lush_forest.jpg" height="128"><img src="docs/assets/000118_597948545_PLMS40_PS7.5_a_lush_forest.jpg" height="128"><img src="docs/assets/000118_597948545_PLMS40_PS7.5_a_lush_forest.jpg" height="128">
 <br>
-<img src="assets/000075_961095192_PLMS40_PS7.5_piles_of_old_books.jpg" height="128"><img src="assets/000075_961095192_PLMS40_PS7.5_piles_of_old_books.jpg" height="128"><img src="assets/000075_961095192_PLMS40_PS7.5_piles_of_old_books.jpg" height="128">
-<img src="assets/000040_527733581_PLMS40_PS7.5_leaves.jpg" height="128"><img src="assets/000040_527733581_PLMS40_PS7.5_leaves.jpg" height="128"><img src="assets/000040_527733581_PLMS40_PS7.5_leaves.jpg" height="128">
+<img src="docs/assets/000075_961095192_PLMS40_PS7.5_piles_of_old_books.jpg" height="128"><img src="docs/assets/000075_961095192_PLMS40_PS7.5_piles_of_old_books.jpg" height="128"><img src="docs/assets/000075_961095192_PLMS40_PS7.5_piles_of_old_books.jpg" height="128">
+<img src="docs/assets/000040_527733581_PLMS40_PS7.5_leaves.jpg" height="128"><img src="docs/assets/000040_527733581_PLMS40_PS7.5_leaves.jpg" height="128"><img src="docs/assets/000040_527733581_PLMS40_PS7.5_leaves.jpg" height="128">
 #### 360 degree images
 ```bash
 imagine --tile-x -w 1024 -h 512 "360 degree equirectangular panorama photograph of the desert"  --upscale
 ```
-<img src="assets/desert_360.jpg" height="128">
+<img src="docs/assets/desert_360.jpg" height="128">
 
 ### Image-to-Image
 Use depth maps for amazing "translations" of existing images.
 
 ```bash
->> imagine --model SD-2.0-depth --init-image girl_with_a_pearl_earring_large.jpg --init-image-strength 0.05  "professional headshot photo of a woman with a pearl earring" -r 4 -w 1024 -h 1024 --steps 50
+>> imagine --init-image girl_with_a_pearl_earring_large.jpg --init-image-strength 0.05  "professional headshot photo of a woman with a pearl earring" -r 4 -w 1024 -h 1024 --steps 50
 ```
 <p float="left">
 <img src="tests/data/girl_with_a_pearl_earring.jpg" width="256"> ➡️ 
-<img src="assets/pearl_depth_1.jpg" width="256">
-<img src="assets/pearl_depth_2.jpg" width="256">
+<img src="docs/assets/pearl_depth_1.jpg" width="256">
+<img src="docs/assets/pearl_depth_2.jpg" width="256">
 </p>
 
 
@@ -289,12 +431,9 @@ Example:
 ### Work with different generation models
 
 <p float="left">
-    <img src="assets/fairytale-treehouse-sd14.jpg" height="256">
-    <img src="assets/fairytale-treehouse-sd15.jpg" height="256">
-    <img src="assets/fairytale-treehouse-sd20.jpg" height="256">
-    <img src="assets/fairytale-treehouse-sd21.jpg" height="256">
-    <img src="assets/fairytale-treehouse-openjourney-v1.jpg" height="256">
-    <img src="assets/fairytale-treehouse-openjourney-v2.jpg" height="256">
+    <img src="docs/assets/fairytale-treehouse-sd15.jpg" height="256">
+    <img src="docs/assets/fairytale-treehouse-openjourney-v1.jpg" height="256">
+    <img src="docs/assets/fairytale-treehouse-openjourney-v2.jpg" height="256">
 </p>
 
 <details>
@@ -323,15 +462,27 @@ You can use `{}` to randomly pull values from lists.  A list of values separated
 
    `imagine "a {lime|blue|silver|aqua} colored dog" -r 4 --seed 0` (note that it generates a dog of each color without repetition)
 
-<img src="assets/000184_0_plms40_PS7.5_a_silver_colored_dog_[generated].jpg" height="200"><img src="assets/000186_0_plms40_PS7.5_a_aqua_colored_dog_[generated].jpg" height="200">
-<img src="assets/000210_0_plms40_PS7.5_a_lime_colored_dog_[generated].jpg" height="200">
-<img src="assets/000211_0_plms40_PS7.5_a_blue_colored_dog_[generated].jpg" height="200">
+<img src="docs/assets/000184_0_plms40_PS7.5_a_silver_colored_dog_[generated].jpg" height="200"><img src="docs/assets/000186_0_plms40_PS7.5_a_aqua_colored_dog_[generated].jpg" height="200">
+<img src="docs/assets/000210_0_plms40_PS7.5_a_lime_colored_dog_[generated].jpg" height="200">
+<img src="docs/assets/000211_0_plms40_PS7.5_a_blue_colored_dog_[generated].jpg" height="200">
 
    `imagine "a {_color_} dog" -r 4 --seed 0` will generate four, different colored dogs. The colors will be pulled from an included 
    phraselist of colors.
     
    `imagine "a {_spaceship_|_fruit_|hot air balloon}. low-poly" -r 4 --seed 0` will generate images of spaceships or fruits or a hot air balloon
 
+<details>
+<summary>Python example</summary>
+
+```python
+from imaginairy.enhancers.prompt_expansion import expand_prompts
+
+my_prompt = "a giant {_animal_}"
+
+expanded_prompts = expand_prompts(n=10, prompt_text=my_prompt, prompt_library_paths=["./prompts"])
+```
+</details>
+
    Credit to [noodle-soup-prompts](https://github.com/WASasquatch/noodle-soup-prompts/) where most, but not all, of the wordlists originate.
 
 ### Generate image captions (via [BLIP](https://github.com/salesforce/BLIP))
@@ -369,8 +520,7 @@ a bowl full of gold bars sitting on a table
  - Prompt metadata saved into image file metadata
  - Have AI generate captions for images `aimg describe <filename-or-url>`
  - Interactive prompt: just run `aimg`
- - finetune your own image model. kind of like dreambooth. Read instructions on ["Concept Training"](docs/concept-training.md) page
-
+ 
 ## How To
 
 For full command line instructions run `aimg --help`
@@ -430,311 +580,21 @@ docker run -it --gpus all -v $HOME/.cache/huggingface:/root/.cache/huggingface -
 
 ## Q&A
 
-**Q**: How do I change the cache directory for where models are stored?
+#### Q: How do I change the cache directory for where models are stored?
 
-**A**: Set the `HUGGINGFACE_HUB_CACHE` environment variable. 
+A: Set the `HUGGINGFACE_HUB_CACHE` environment variable. 
 
+#### Q: How do I free up disk space?
 
-## ChangeLog
+A: The AI models are cached in `~/.cache/` (or `HUGGINGFACE_HUB_CACHE`). To delete the cache remove the following folders:
+ - ~/.cache/imaginairy
+ - ~/.cache/clip
+ - ~/.cache/torch
+ - ~/.cache/huggingface
 
-**12.0.1**
-- fix: use correct device for depth images on mps. Fixes #300
-
-**12.0.0**
-
-- 🎉 feature: add "detail" control mode.  Add details to an image. Great for upscaling an image.
-- 🎉 feature: add "edit" control mode.  Edit images using text instructions with any SD 1.5 based model. Similar to instructPix2Pix.
-- 🎉 feature: add "shuffle" control mode. Image is generated from elements of control image. Similar to style transfer.
-- 🎉 feature: upgrade to [controlnet 1.1](https://github.com/lllyasviel/ControlNet-v1-1-nightly)
-- 🎉 fix: controlnet now works with all SD 1.5 based models
-- feature: add openjourney-v4
-- fix: raw control images are now properly loaded. fixes #296
-- fix: filenames start numbers after latest image, even if some previous images were deleted
-
-**11.1.1**
-- fix: fix globbing bug with input image path handling
-- fix: changed sample to True to generate caption using blip model
-
-**11.1.0**
-
-- docs: add some example use cases
-- feature: add art-scene, desktop-background, interior-style, painting-style phraselists
-- 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
-- fix: remove upscaler tiling message
-- fix: improve k-diffusion sampler schedule. significantly improves image quality of default sampler
-- fix: img2img was broken for all samplers except plms and ddim when init image strength was >~0.25
-
-**10.2.0**
- - feature: input raw control images (a pose, canny map, depth map, etc) directly using `--control-image-raw`
-   This is opposed to current behavior of extracting the control signal from an input image via `--control-image`
- - feature: `aimg model-list` command lists included models
- - feature: system memory added to `aimg system-info` command
- - feature: add `--fix-faces` options to `aimg upscale` command
- - fix: add missing metadata attributes to generated images
- - fix: image composition step was producing unnecessarily blurry images 
- - refactor: split `aimg` cli code into multiple files
- - docs: pypi docs now link properly to github automatically
-
-**10.1.0**
-- feature: 🎉 ControlNet integration!  Control the structure of generated images.
-- feature: `aimg colorize` attempts to use controlnet to colorize images
-- feature: `--caption-text` command adds text at the bottom left of an image
-
-**10.0.1**
-- fix: `edit` was broken
-
-**10.0.0**
-
-- feature: 🎉🎉 Make large images while retaining composition. Try `imagine "a flower" -w 1920 -h 1080`
-- fix: create compilations directory automatically
-- perf: sliced encoding of images to latents (removes memory bottleneck)
-- perf: use Silu for performance improvement over nonlinearity
-- perf: `xformers` added as a dependency for linux and windows.  Gives a nice speed boost.
-- perf: sliced attention now runs on MacOS. A typo prevented that from happening previously.
-- perf: sliced latent decoding - now possible to make much bigger images. 3310x3310 on 11 GB GPU.
-
-**9.0.2**
-- fix: edit interface was broken
-
-**9.0.1**
-- fix: use entry_points for windows since setup.py scripts doesn't work on windows [#239](https://github.com/brycedrennan/imaginAIry/issues/239)
-
-**9.0.0**
-
-- perf: cli now has minimal overhead such that `aimg --help` runs in ~650ms instead of ~3400ms
-- feature: `edit` and `imagine` commands now accept multiple images (which they will process separately).  This allows 
-batch editing of images as requested in [#229](https://github.com/brycedrennan/imaginAIry/issues/229)
-- refactor: move `--surprise-me` to its own subcommand `edit-demo`
-- feature: allow selection of output image format with `--output-file-extension`
-- docs: make training fail on MPS platform with useful error message
-- docs: add directions on how to change model cache path
-
-**8.3.1**
-- fix: init-image-strength type
-
-**8.3.0**
-- feature: create `gifs` or `mp4s` from any images made in a single run with `--compilation-anim gif`
-- feature: create a series of images or edits by iterating over a parameter with the `--arg-schedule` argument
-- feature: `openjourney-v1` and `openjourney-v2` models added. available via `--model openjourney-v2`
-- feature: add upscale command line function: `aimg upscale`
-- feature: `--gif` option will create a gif showing the generation process for a single image
-- feature: `--compare-gif` option will create a comparison gif for any image edits
-- fix: tile mode was broken since latest perf improvements
-
-**8.2.0**
-- feature: added `aimg system-info` command to help debug issues
-
-**8.1.0**
-- feature: some memory optimizations and documentation
-- feature: surprise-me improvements
-- feature: image sizes can now be multiples of 8 instead of 64. Inputs will be silently rounded down.
-- feature: cleaned up `aimg` shell logs  
-- feature: auto-regen for unsafe images
-- fix: make blip filename windows compatible
-- fix: make captioning work with alpha pngs
- 
-**8.0.5**
-- fix: bypass huggingface cache retrieval bug
-
-**8.0.4**
-- fix: limit attention slice size on MacOS machines with 64gb (#175)
-
-**8.0.3**
-- fix: use python 3.7 compatible lru_cache
-- fix: use windows compatible filenames
-
-**8.0.2**
-- fix: hf_hub_download() got an unexpected keyword argument 'token'
-
-**8.0.1**
-- fix: spelling mistake of "surprise"
-
-**8.0.0**
-- feature: 🎉 edit images with instructions alone!
-- feature: when editing an image add `--gif` to create a comparision gif
-- feature: `aimg edit --surprise-me --gif my-image.jpg` for some fun pre-programmed edits
-- feature: prune-ckpt command also removes the non-ema weights
-
-**7.6.0**
-- fix: default model config was broken
-- feature: print version with `--version`
-- feature: ability to load safetensors
-- feature:  🎉 outpainting. Examples: `--outpaint up10,down300,left50,right50` or `--outpaint all100` or `--outpaint u100,d200,l300,r400`
-
-**7.4.3**
-- fix: handle old pytorch lightning imports with a graceful failure (fixes #161)
-- fix: handle failed image generations better (fixes #83)
-
-**7.4.2**
-- fix: run face enhancement on GPU for 10x speedup
-
-**7.4.1**
-- fix: incorrect config files being used for non-1.0 models
-
-**7.4.0**
-- feature: 🎉 finetune your own image model. kind of like dreambooth. Read instructions on ["Concept Training"](docs/concept-training.md) page
-- feature: image prep command. crops to face or other interesting parts of photo
-- fix: back-compat for hf_hub_download
-- feature: add prune-ckpt command
-- feature: allow specification of model config file
-
-**7.3.0**
-- feature: 🎉 depth-based image-to-image generations (and inpainting) 
-- fix: k_euler_a produces more consistent images per seed (randomization respects the seed again)
-
-**7.2.0**
-- feature: 🎉 tile in a single dimension ("x" or "y").  This enables, with a bit of luck, generation of 360 VR images.
-Try this for example: `imagine --tile-x -w 1024 -h 512 "360 degree equirectangular panorama photograph of the mountains"  --upscale`
-
-**7.1.1**
-- fix: memory/speed regression introduced in 6.1.0
-- fix: model switching now clears memory better, thus avoiding out of memory errors
-
-**7.1.0**
-- feature: 🎉 Stable Diffusion 2.1.  Generated people are no longer (completely) distorted. 
-Use with `--model SD-2.1` or `--model SD-2.0-v` 
-
-**7.0.0**
-- feature: negative prompting.  `--negative-prompt` or `ImaginePrompt(..., negative_prompt="ugly, deformed, extra arms, etc")`
-- feature: a default negative prompt is added to all generations. Images in SD-2.0 don't look bad anymore. Images in 1.5 look improved as well.
-
-**6.1.2**
-- fix: add back in memory-efficient algorithms
-
-**6.1.1**
-- feature: xformers will be used if available (for faster generation)
-- fix: version metadata was broken
-
-**6.1.0**
-- feature: use different default steps and image sizes depending on sampler and model selceted
-- fix: #110 use proper version in image metadata
-- refactor: samplers all have their own class that inherits from ImageSampler
-- feature: 🎉🎉🎉 Stable Diffusion 2.0
-  - `--model SD-2.0` to use (it makes worse images than 1.5 though...) 
-  - Tested on macOS and Linux
-  - All samplers working for new 512x512 model
-  - New inpainting model working
-  - 768x768 model working for all samplers except PLMS (`--model SD-2.0-v `)
-
-**5.1.0**
-- feature: add progress image callback
-
-**5.0.1**
-- fix: support larger images on M1. Fixes #8
-- fix: support CPU generation by disabling autocast on CPU. Fixes #81
-
-**5.0.0**
-- feature: 🎉 inpainting support using new inpainting model from RunwayML. It works really well! By default, the 
-inpainting model will automatically be used for any image-masking task 
-- feature: 🎉 new default sampler makes image generation more than twice as fast
-- feature: added `DPM++ 2S a` and `DPM++ 2M` samplers.
-- feature: improve progress image logging
-- fix: fix bug with `--show-work`. fixes #84
-- fix: add workaround for pytorch bug affecting macOS users using the new `DPM++ 2S a` and `DPM++ 2M` samplers.
-- fix: add workaround for pytorch mps bug affecting `k_dpm_fast` sampler. fixes #75
-- fix: larger image sizes now work on macOS. fixes #8
-
-**4.1.0**
- - feature: allow dynamic switching between models/weights `--model SD-1.5` or `--model SD-1.4` or `--model path/my-custom-weights.ckpt`)
- - feature: log total progress when generating images (image X out of Y)
-
-**4.0.0**
- - feature: stable diffusion 1.5 (slightly improved image quality)
- - feature: dilation and erosion of masks
- 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
- - refactor: consolidates img2img/txt2img code. consolidates schedules. consolidates masking
- - ci: minor logging improvements
-
-**3.0.1**
- - fix: k-samplers were broken
-
-**3.0.0**
- - feature: improved safety filter
-
-**2.4.0**
- - 🎉 feature: prompt expansion
- - 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"
-
-**1.3.0**
- - added prompt based image editing. Example: "fruit => gold coins"
- - test coverage improved
-
-**1.2.0**
- - allow urls as init-images
-
-**previous**
- - img2img actually does # of steps you specify  
- - performance optimizations
- - numerous other changes
 
 
 ## Not Supported
- - a GUI. this is a python library
  - exploratory features that don't work well