Update versions (#99)

- auto:
ca3e5519e8
- hlky:
1fd28eed1e
- lstein:
b40bfb5116

.github/workflows/stale.yml

@@ -18,3 +18,4 @@ jobs:
           days-before-pr-stale: 14
           days-before-issue-close: 7
           days-before-pr-close: 7
+          ignore-updates:  true

.gitignore

@@ -1,2 +1,3 @@
 /dev
 /.devcontainer
+embeddings/*

LICENSE

@@ -86,4 +86,11 @@ administration of justice, law enforcement, immigration or asylum
 processes, such as predicting an individual will commit fraud/crime
 commitment (e.g. by text profiling, drawing causal relationships between
 assertions made in documents, indiscriminate and arbitrarily-targeted
-use).
+use).
+
+
+
+By using this software, you also agree to the following licenses:
+https://github.com/CompVis/stable-diffusion/blob/main/LICENSE
+https://github.com/TencentARC/GFPGAN/blob/master/LICENSE
+https://github.com/xinntao/Real-ESRGAN/blob/master/LICENSE

README.md

@@ -16,6 +16,7 @@ This repository provides multiple UIs for you to play around with stable diffusi
 - Loopback, prompt weighting, prompt matrix, X/Y plot
 - Live preview of the generated images.
 - Highly optimized 4GB GPU support, or even CPU only!
+- Textual inversion allows you to use pretrained textual inversion embeddings
 - [Full feature list here](https://github.com/AUTOMATIC1111/stable-diffusion-webui-feature-showcase)
 
 | Text to image                                                                                              | Image to image                                                                                             | Extras                                                                                                     |

docker-compose.yml

@@ -38,8 +38,9 @@ services:
       - *v1
       - *v2
       - ./services/AUTOMATIC1111/config.json:/stable-diffusion-webui/config.json
+      - ./embeddings:/stable-diffusion-webui/embeddings
     environment:
-      - CLI_ARGS=--medvram
+      - CLI_ARGS=--allow-code --medvram
 
   automatic1111-cpu:
     <<: *automatic

services/AUTOMATIC1111/Dockerfile

@@ -3,14 +3,17 @@
 FROM alpine/git:2.36.2 as download
 
 
-RUN git clone --depth 1 https://github.com/CompVis/stable-diffusion.git repositories/stable-diffusion
-RUN git clone --depth 1 https://github.com/sczhou/CodeFormer.git repositories/CodeFormer
-RUN git clone --depth 1 https://github.com/salesforce/BLIP.git repositories/BLIP
+RUN git clone https://github.com/CompVis/stable-diffusion.git repositories/stable-diffusion && cd repositories/stable-diffusion && git reset --hard 69ae4b35e0a0f6ee1af8bb9a5d0016ccb27e36dc
+
+RUN git clone https://github.com/sczhou/CodeFormer.git repositories/CodeFormer && cd repositories/CodeFormer && git reset --hard c5b4593074ba6214284d6acd5f1719b6c5d739af
+RUN git clone https://github.com/salesforce/BLIP.git repositories/BLIP && cd repositories/BLIP && git reset --hard 48211a1594f1321b00f14c9f7a5b4813144b2fb9
+RUN git clone https://github.com/Hafiidz/latent-diffusion.git repositories/latent-diffusion && cd repositories/latent-diffusion && git reset --hard abf33e7002d59d9085081bce93ec798dcabd49af
 
 RUN <<EOF
 # because taming-transformers is huge
 git config --global http.postBuffer 1048576000
-git clone --depth 1 https://github.com/CompVis/taming-transformers.git repositories/taming-transformers
+git clone https://github.com/CompVis/taming-transformers.git repositories/taming-transformers
+git reset --hard 24268930bf1dce879235a7fddd0b2355b84d7ea6
 rm -rf repositories/taming-transformers/data repositories/taming-transformers/assets
 EOF
 
@@ -44,12 +47,13 @@ RUN pip install --prefer-binary --no-cache-dir -r ${ROOT}/repositories/CodeForme
 # Note: don't update the sha of previous versions because the install will take forever
 # instead, update the repo state in a later step
 
-ARG SHA=dd911a47b3c3313b3938b700eb26cbd5bb3e1c95
+ARG SHA=ca3e5519e8b6dc020c5e7ae508738afb5dc6f3ec
 RUN <<EOF
 cd stable-diffusion-webui
 git pull --rebase
 git reset --hard ${SHA}
 pip install --prefer-binary --no-cache-dir -r requirements.txt
+pip install --prefer-binary --no-cache-dir -r requirements_versions.txt
 EOF
 
 RUN pip install --prefer-binary -U --no-cache-dir opencv-python-headless
@@ -57,11 +61,15 @@ RUN pip install --prefer-binary -U --no-cache-dir opencv-python-headless
 ENV TRANSFORMERS_CACHE=/cache/transformers TORCH_HOME=/cache/torch CLI_ARGS=""
 
 COPY . /docker
-RUN chmod +x /docker/mount.sh && python3 /docker/info.py ${ROOT}/modules/ui.py
-
+RUN <<EOF
+chmod +x /docker/mount.sh && python3 /docker/info.py ${ROOT}/modules/ui.py
+# hackiest of hacks, change default cache dir of clip #88
+# https://github.com/openai/CLIP/blob/d50d76daa670286dd6cacf3bcd80b5e4823fc8e1/clip/clip.py#L94
+sed -i -- 's/download_root: str = None/download_root: str = "\/cache\/weights"/' /opt/conda/lib/python3.8/site-packages/clip/clip.py
+EOF
 
 WORKDIR ${WORKDIR}
 EXPOSE 7860
 # run, -u to not buffer stdout / stderr
 CMD /docker/mount.sh && \
-  python3 -u ../../webui.py --listen --port 7860 --hide-ui-dir-config --ckpt-dir /cache/custom-models --ckpt /cache/models/model.ckpt ${CLI_ARGS}
+  python3 -u ../../webui.py --listen --port 7860 --hide-ui-dir-config --ckpt-dir /cache/custom-models --ckpt /cache/models/model.ckpt --gfpgan-model /cache/models/GFPGANv1.3.pth ${CLI_ARGS}

services/AUTOMATIC1111/config.json

@@ -9,50 +9,71 @@
   "font": "DejaVuSans.ttf",
   "__WARNING__": "DON'T CHANGE ANYTHING BEFORE THIS",
 
-  "samples_filename_format": "",
-  "outdir_grids": "",
-  "save_to_dirs": false,
-  "grid_save_to_dirs": false,
-  "save_to_dirs_prompt_len": 10,
-  "samples_save": true,
-  "samples_format": "png",
-  "grid_save": true,
-  "return_grid": true,
-  "grid_format": "png",
-  "grid_extended_filename": false,
-  "grid_only_if_multiple": true,
-  "n_rows": -1,
-  "jpeg_quality": 80,
-  "export_for_4chan": true,
-  "enable_pnginfo": true,
   "add_model_hash_to_info": false,
+  "code_former_weight": 0.5,
+  "directories_filename_pattern": "",
+  "directories_max_prompt_words": 8,
+  "enable_batch_seeds": true,
   "enable_emphasis": true,
-  "save_txt": false,
+  "enable_pnginfo": true,
+  "enable_quantization": false,
   "ESRGAN_tile": 192,
   "ESRGAN_tile_overlap": 8,
-  "random_artist_categories": [],
-  "upscale_at_full_resolution_padding": 16,
-  "show_progressbar": true,
-  "show_progress_every_n_steps": 7,
-  "multiple_tqdm": true,
+  "export_for_4chan": true,
   "face_restoration_model": null,
-  "code_former_weight": 0.5,
-  "save_images_before_face_restoration": false,
   "face_restoration_unload": false,
-  "interrogate_keep_models_in_memory": false,
-  "interrogate_use_builtin_artists": true,
-  "interrogate_clip_num_beams": 1,
-  "interrogate_clip_min_length": 24,
-  "interrogate_clip_max_length": 48,
-  "interrogate_clip_dict_limit": 1500.0,
-  "samples_filename_pattern": "",
-  "directories_filename_pattern": "",
-  "save_selected_only": false,
   "filter_nsfw": false,
+  "grid_extended_filename": false,
+  "grid_format": "png",
+  "grid_only_if_multiple": true,
+  "grid_save": true,
+  "grid_save_to_dirs": false,
   "img2img_color_correction": false,
   "img2img_fix_steps": false,
-  "enable_quantization": false,
-  "enable_batch_seeds": true,
+  "interrogate_clip_dict_limit": 1500,
+  "interrogate_clip_max_length": 48,
+  "interrogate_clip_min_length": 24,
+  "interrogate_clip_num_beams": 1,
+  "interrogate_keep_models_in_memory": false,
+  "interrogate_use_builtin_artists": true,
+  "jpeg_quality": 80,
+  "js_modal_lightbox": true,
+  "js_modal_lightbox_initialy_zoomed": true,
+  "ldsr_post_down": 1,
+  "ldsr_pre_down": 1,
+  "ldsr_steps": 30,
   "memmon_poll_rate": 8,
-  "sd_model_checkpoint": null
+  "multiple_tqdm": true,
+  "n_rows": -1,
+  "outdir_grids": "",
+  "random_artist_categories": [],
+  "realesrgan_enabled_models": [
+    "Real-ESRGAN 4x plus",
+    "Real-ESRGAN 4x plus anime 6B",
+    "Real-ESRGAN 2x plus",
+    "Real-ESRGAN AnimeVideo",
+    "Real-ESRGAN General WDN x4x3",
+    "Real-ESRGAN General x4x3"
+  ],
+  "return_grid": true,
+  "samples_filename_format": "",
+  "samples_filename_pattern": "",
+  "samples_format": "png",
+  "samples_log_stdout": false,
+  "samples_save": true,
+  "save_images_before_color_correction": false,
+  "save_images_before_face_restoration": false,
+  "save_selected_only": false,
+  "save_to_dirs": false,
+  "save_to_dirs_prompt_len": 10,
+  "save_txt": false,
+  "sd_model_checkpoint": null,
+  "show_progress_every_n_steps": 7,
+  "show_progressbar": true,
+  "SWIN_tile": 192,
+  "SWIN_tile_overlap": 8,
+  "upscale_at_full_resolution_padding": 16,
+  "upscaler_for_hires_fix": null,
+  "upscaler_for_img2img": null,
+  "use_original_name_batch": false
 }

services/AUTOMATIC1111/mount.sh

@@ -4,8 +4,9 @@ set -e
 
 declare -A MODELS
 
-MODELS["${WORKDIR}/models/ldm/stable-diffusion-v1/model.ckpt"]=model.ckpt
 MODELS["${ROOT}/GFPGANv1.3.pth"]=GFPGANv1.3.pth
+MODELS["${WORKDIR}/repositories/latent-diffusion/experiments/pretrained_models/model.chkpt"]=LDSR.ckpt
+MODELS["${WORKDIR}/repositories/latent-diffusion/experiments/pretrained_models/project.yaml"]=LDSR.yaml
 
 MODELS_DIR=/cache/models
 

services/download/download.sh

@@ -4,6 +4,13 @@ set -Eeuo pipefail
 
 mkdir -p /cache/torch /cache/transformers /cache/weights /cache/models /cache/custom-models
 
+cat <<EOF
+By using this software, you agree to the following licenses:
+https://github.com/CompVis/stable-diffusion/blob/main/LICENSE
+https://github.com/TencentARC/GFPGAN/blob/master/LICENSE
+https://github.com/xinntao/Real-ESRGAN/blob/master/LICENSE
+EOF
+
 echo "Downloading, this might take a while..."
 
 aria2c --input-file /docker/links.txt --dir /cache/models --continue

services/hlky/Dockerfile

@@ -21,12 +21,11 @@ conda env update --file environment.yaml -n base
 conda clean -a -y
 EOF
 
+
 # Note: don't update the sha of previous versions because the install will take forever
 # instead, update the repo state in a later step
-ARG BRANCH=master
-# ARG SHA=833a91047df999302f699637768741cecee9c37b
-# ARG BRANCH=dev
-ARG SHA=17748cbc9c34df44d0381c42e4f0fe1903089438
+# ARG BRANCH=master  SHA=d0bb60a139d60e6c2b9be4e18e0e29a86aa5af59
+ARG BRANCH=dev SHA=1fd28eed1ebc3aa04b9b00e2a899f3bf07f64bdc
 RUN <<EOF
 cd stable-diffusion
 git fetch
@@ -36,27 +35,16 @@ conda env update --file environment.yaml -n base
 conda clean -a -y
 EOF
 
-RUN pip uninstall transformers -y && pip install -U --no-cache-dir pyperclip transformers==4.22
-
-# Latent diffusion
-RUN <<EOF
-git clone --depth 1 https://github.com/Hafiidz/latent-diffusion.git
-cd latent-diffusion
-# hacks all the way down
-mv ldm ldm_latent &&
-sed -i -- 's/from ldm/from ldm_latent/g' *.py
-# dont forget to update the yaml!!
-EOF
-
+RUN pip install -U --no-cache-dir pyperclip
 
 # add info
 COPY . /docker/
 RUN python /docker/info.py /stable-diffusion/frontend/frontend.py && chmod +x /docker/mount.sh
 
 WORKDIR /stable-diffusion
-ENV TRANSFORMERS_CACHE=/cache/transformers TORCH_HOME=/cache/torch PYTHONPATH="${PYTHONPATH}:/stable-diffusion" CLI_ARGS=""
+ENV TRANSFORMERS_CACHE=/cache/transformers TORCH_HOME=/cache/torch PYTHONPATH="${PYTHONPATH}:${PWD}" CLI_ARGS=""
 EXPOSE 7860
 # run, -u to not buffer stdout / stderr
 CMD /docker/mount.sh && \
-  python3 -u scripts/webui.py --outdir /output --ckpt /cache/models/model.ckpt --ldsr-dir /latent-diffusion --inbrowser ${CLI_ARGS}
-# STREAMLIT_SERVER_PORT=7860 python -m streamlit run scripts/webui_streamlit.py
+  python3 -u scripts/webui.py --outdir /output --ckpt /cache/models/model.ckpt ${CLI_ARGS}
+  # STREAMLIT_SERVER_PORT=7860 python -m streamlit run scripts/webui_streamlit.py  --theme.base dark

services/hlky/mount.sh

@@ -9,8 +9,8 @@ ROOT=/stable-diffusion/src
 MODELS["${ROOT}/gfpgan/experiments/pretrained_models/GFPGANv1.3.pth"]=GFPGANv1.3.pth
 MODELS["${ROOT}/realesrgan/experiments/pretrained_models/RealESRGAN_x4plus.pth"]=RealESRGAN_x4plus.pth
 MODELS["${ROOT}/realesrgan/experiments/pretrained_models/RealESRGAN_x4plus_anime_6B.pth"]=RealESRGAN_x4plus_anime_6B.pth
-MODELS["/latent-diffusion/experiments/pretrained_models/model.ckpt"]=LDSR.ckpt
-# MODELS["/latent-diffusion/experiments/pretrained_models/project.yaml"]=LDSR.yaml
+MODELS["${ROOT}/latent-diffusion/experiments/pretrained_models/model.ckpt"]=LDSR.ckpt
+MODELS["${ROOT}/latent-diffusion/experiments/pretrained_models/project.yaml"]=LDSR.yaml
 
 MODELS_DIR=/cache/models
 
@@ -25,14 +25,9 @@ for path in "${!MODELS[@]}"; do
   fi
 done
 
-# hack for latent-diffusion
-if test -f "${MODELS_DIR}/LDSR.yaml"; then
-  sed 's/ldm\./ldm_latent\./g' "${MODELS_DIR}/LDSR.yaml" >/latent-diffusion/experiments/pretrained_models/project.yaml
-fi
-
 # force facexlib cache
 mkdir -p /cache/weights/ /stable-diffusion/gfpgan/
 ln -sf /cache/weights/ /stable-diffusion/gfpgan/
 
 # streamlit config
-ln -sf /docker/webui_streamlit.yaml /stable-diffusion/configs/webui/webui_streamlit.yaml
+ln -sf /docker/userconfig_streamlit.yaml /stable-diffusion/configs/webui/userconfig_streamlit.yaml

services/hlky/userconfig_streamlit.yaml

@@ -0,0 +1,8 @@
+general:
+  outdir: /outputs
+  default_model: "Stable Diffusion v1.4"
+  default_model_path: /cache/models/model.ckpt
+  outdir_txt2img: /outputs/txt2img-samples
+  outdir_img2img: /outputs/img2img-samples
+  optimized: True
+  optimized_turbo: True

services/hlky/webui_streamlit.yaml

@@ -1,155 +0,0 @@
-# UI defaults configuration file. It is automatically loaded if located at configs/webui/webui_streamlit.yaml.
-general:
-  gpu: 0
-  outdir: /outputs
-
-  default_model: "Stable Diffusion v1.4"
-  default_model_config: "configs/stable-diffusion/v1-inference.yaml"
-  default_model_path: "/cache/models/model.ckpt"
-  fp:
-    name:
-  GFPGAN_dir: "./src/gfpgan"
-  RealESRGAN_dir: "./src/realesrgan"
-  RealESRGAN_model: "RealESRGAN_x4plus"
-  outdir_txt2img: /outputs/txt2img-samples
-  outdir_img2img: /outputs/img2img-samples
-  gfpgan_cpu: False
-  esrgan_cpu: False
-  extra_models_cpu: False
-  extra_models_gpu: False
-  save_metadata: True
-  save_format: "png"
-  skip_grid: False
-  skip_save: False
-  grid_format: "jpg:95"
-  n_rows: -1
-  no_verify_input: False
-  no_half: False
-  use_float16: False
-  precision: "autocast"
-  optimized: False
-  optimized_turbo: True
-  optimized_config: "optimizedSD/v1-inference.yaml"
-  update_preview: True
-  update_preview_frequency: 5
-
-txt2img:
-  prompt:
-  height: 512
-  width: 512
-  cfg_scale: 7.5
-  seed: ""
-  batch_count: 1
-  batch_size: 1
-  sampling_steps: 30
-  default_sampler: "k_euler"
-  separate_prompts: False
-  update_preview: True
-  update_preview_frequency: 5
-  normalize_prompt_weights: True
-  save_individual_images: True
-  save_grid: True
-  group_by_prompt: True
-  save_as_jpg: False
-  use_GFPGAN: False
-  use_RealESRGAN: False
-  RealESRGAN_model: "RealESRGAN_x4plus"
-  variant_amount: 0.0
-  variant_seed: ""
-  write_info_files: True
-
-txt2vid:
-  default_model: "CompVis/stable-diffusion-v1-4"
-  custom_models_list:
-    [
-      "CompVis/stable-diffusion-v1-4",
-      "naclbit/trinart_stable_diffusion_v2",
-      "hakurei/waifu-diffusion",
-      "osanseviero/BigGAN-deep-128",
-    ]
-  prompt:
-  height: 512
-  width: 512
-  cfg_scale: 7.5
-  seed: ""
-  batch_count: 1
-  batch_size: 1
-  sampling_steps: 30
-  num_inference_steps: 200
-  default_sampler: "k_euler"
-  scheduler_name: "klms"
-  separate_prompts: False
-  update_preview: True
-  update_preview_frequency: 5
-  dynamic_preview_frequency: True
-  normalize_prompt_weights: True
-  save_individual_images: True
-  save_video: True
-  group_by_prompt: True
-  write_info_files: True
-  do_loop: False
-  save_as_jpg: False
-  use_GFPGAN: False
-  use_RealESRGAN: False
-  RealESRGAN_model: "RealESRGAN_x4plus"
-  variant_amount: 0.0
-  variant_seed: ""
-  beta_start: 0.00085
-  beta_end: 0.012
-  beta_scheduler_type: "linear"
-  max_frames: 1000
-
-img2img:
-  prompt:
-  sampling_steps: 30
-  # Adding an int to toggles enables the corresponding feature.
-  # 0: Create prompt matrix (separate multiple prompts using |, and get all combinations of them)
-  # 1: Normalize Prompt Weights (ensure sum of weights add up to 1.0)
-  # 2: Loopback (use images from previous batch when creating next batch)
-  # 3: Random loopback seed
-  # 4: Save individual images
-  # 5: Save grid
-  # 6: Sort samples by prompt
-  # 7: Write sample info files
-  # 8: jpg samples
-  # 9: Fix faces using GFPGAN
-  # 10: Upscale images using Real-ESRGAN
-  sampler_name: "k_euler"
-  denoising_strength: 0.45
-  # 0: Keep masked area
-  # 1: Regenerate only masked area
-  mask_mode: 0
-  mask_restore: False
-  # 0: Just resize
-  # 1: Crop and resize
-  # 2: Resize and fill
-  resize_mode: 0
-  # Leave blank for random seed:
-  seed: ""
-  ddim_eta: 0.0
-  cfg_scale: 7.5
-  batch_count: 1
-  batch_size: 1
-  height: 512
-  width: 512
-  # Textual inversion embeddings file path:
-  fp: ""
-  loopback: True
-  random_seed_loopback: True
-  separate_prompts: False
-  update_preview: True
-  update_preview_frequency: 5
-  normalize_prompt_weights: True
-  save_individual_images: True
-  save_grid: True
-  group_by_prompt: True
-  save_as_jpg: False
-  use_GFPGAN: False
-  use_RealESRGAN: False
-  RealESRGAN_model: "RealESRGAN_x4plus"
-  variant_amount: 0.0
-  variant_seed: ""
-  write_info_files: True
-
-gfpgan:
-  strength: 100

services/lstein/Dockerfile

@@ -14,17 +14,19 @@ RUN apt-get update && apt install fonts-dejavu-core rsync gcc -y && apt-get clea
 
 
 RUN <<EOF
-git clone https://github.com/lstein/stable-diffusion.git
+git clone https://github.com/invoke-ai/InvokeAI.git stable-diffusion
 cd stable-diffusion
-git reset --hard e994073b5bdfa3c77313681c5944be1544eb65b6
+git reset --hard a1739a73b48bfe98b6abcb67f5a0197a9ad270e0
 sed -i -- 's/python=3.8.5/python=3.9/g' environment.yaml
 conda env update --file environment.yaml -n base
 conda clean -a -y
 EOF
 
 
-ARG BRANCH=development SHA=50d607ffea3734072a80e38b09ba0c3758af5d40
-# ARG BRANCH=main SHA=89da371f4841f7e05da5a1672459d700c3920784
+ARG BRANCH=development SHA=b40bfb5116b7fc618f78a0d152005ceb46153443
+# this breaks on generation:
+# there is a new UI anyway, but it is not by any means ready.
+# ARG BRANCH=development SHA=bdbc76fcd4bd3362312dc91b087d9af66de423b1
 RUN <<EOF
 cd stable-diffusion
 git fetch
@@ -35,7 +37,7 @@ conda env update --file environment.yaml -n base
 conda clean -a -y
 EOF
 
-RUN pip uninstall opencv-python -y && pip install --prefer-binary --force-reinstall --no-cache-dir opencv-python-headless transformers==4.22
+RUN pip uninstall opencv-python -y && pip install --prefer-binary --force-reinstall --no-cache-dir opencv-python-headless transformers==4.19.2
 
 COPY . /docker/
 RUN <<EOF
@@ -51,4 +53,4 @@ EXPOSE 7860
 
 CMD /docker/mount.sh && \
   python3 -u scripts/dream.py --outdir /output --web --host 0.0.0.0 --port 7860 ${CLI_ARGS}
-  #python3 -u backend/server.py
+# python3 -u backend/server.py --host 0.0.0.0 --port 9090

services/lstein/mount.sh

@@ -7,12 +7,11 @@ ROOT=/stable-diffusion
 mkdir -p "${ROOT}/models/ldm/stable-diffusion-v1/"
 ln -sf /cache/models/model.ckpt "${ROOT}/models/ldm/stable-diffusion-v1/model.ckpt"
 
-if test -f /cache/models/GFPGANv1.3.pth; then
-  base="${ROOT}/src/gfpgan/experiments/pretrained_models/"
-  mkdir -p "${base}"
-  ln -sf /cache/models/GFPGANv1.3.pth "${base}/GFPGANv1.3.pth"
-  echo "Mounted GFPGANv1.3.pth"
-fi
+base="${ROOT}/src/gfpgan/experiments/pretrained_models/"
+mkdir -p "${base}"
+# TODO: "real" GFPGANv1.4.pth
+ln -sf /cache/models/GFPGANv1.3.pth "${base}/GFPGANv1.4.pth"
+echo "Mounted GFPGANv1.3.pth"
 
 # facexlib
 FACEX_WEIGHTS=/opt/conda/lib/python3.9/site-packages/facexlib/weights