diff --git a/README.md b/README.md index 7fca8b8..c1bbe36 100644 --- a/README.md +++ b/README.md @@ -7,12 +7,12 @@ This is a Diffusers-based script for running Stable Diffusion on [Modal](https:/ ## Features 1. Image generation using txt2img -![](assets/20230902_tile_imgs.png) + ![](assets/20230902_tile_imgs.png) 2. Upscaling -| Before upscaling | After upscaling | -| ---- | ---- | +| Before upscaling | After upscaling | +| ---------------------------------------------------------------- | ---------------------------------------------------------------- | | | | ## Requirements @@ -105,9 +105,7 @@ controlnets: repo_id: lllyasviel/control_v11f1e_sd15_tile ``` -Use a model configured for Diffusers, such as the one found in [this repository](https://huggingface.co/stabilityai/stable-diffusion-2-1). - -Files in safetensor format shared by Civitai etc. need to be converted (you can do so with a script in the diffusers official repository). +Use a model configured for Diffusers, such as the one found in [this repository](https://huggingface.co/stabilityai/stable-diffusion-2-1). Files in safetensor format shared by Civitai etc. need to be converted (you can do so with a script in the diffusers official repository). [https://github.com/huggingface/diffusers/blob/main/scripts/convert_original_stable_diffusion_to_diffusers.py](https://github.com/huggingface/diffusers/blob/main/scripts/convert_original_stable_diffusion_to_diffusers.py) @@ -119,6 +117,15 @@ python ./diffusers/scripts/convert_original_stable_diffusion_to_diffusers.py --f --device='cuda:0' ``` +LoRA and Textual Inversion don't require any conversion and can directly use safetensors files. Add the download link to config.yml as below. + +``` +# Example +loras: + - name: lora_name.safetensors # Specify the LoRA file name. Any name is fine, but the extension `.safetensors` is required. + download_url: download_link_here # Specify the download link for the safetensor file. +``` + ### 4. Setting prompts Set the prompt to Makefile. diff --git a/README_ja.md b/README_ja.md index c299b50..60e491b 100644 --- a/README_ja.md +++ b/README_ja.md @@ -10,8 +10,8 @@ 2. アップスケーラーとControlNet Tileを利用した高解像度な画像を生成することができます。 -| ベース画像 | アップスケール後 | -| ---- | ---- | +| ベース画像 | アップスケール後 | +| ---------------------------------------------------------------- | ---------------------------------------------------------------- | | | | 3. その他、LoRAとTextual inversionを利用できます。 @@ -105,12 +105,19 @@ controlnets: repo_id: lllyasviel/control_v11f1e_sd15_tile ``` -ModelとVAEは[こちらのリポジトリ](https://huggingface.co/stabilityai/stable-diffusion-2-1)にあるような、Diffusersのために構成されたモデルを利用します。 - -Civitaiなどで共有されているsafetensors形式のファイルは変換が必要です（diffusersの公式リポジトリにあるスクリプトで変換できます）。 +ModelとVAEは[こちらのリポジトリ](https://huggingface.co/stabilityai/stable-diffusion-2-1)にあるような、Diffusersのために構成されたモデルを利用します。Civitaiなどで共有されているsafetensors形式のファイルは変換が必要です（diffusersの公式リポジトリにあるスクリプトで変換できます）。 [変換スクリプト](https://github.com/huggingface/diffusers/blob/main/scripts/convert_original_stable_diffusion_to_diffusers.py) +LoRAとTextual Inversionは変換不要で、safetensorsファイルをそのまま利用できます。 + +``` +# 設定例 +loras: + - name: mecha.safetensors # ファイル名を指定。任意の名前で良いが、拡張子`.safetensors`は必須。 + download_url: https://civitai.com/api/download/models/150907?type=Model&format=SafeTensor # ダウンロードリンクを指定 +``` + ``` # 変換スクリプトの使用例 python ./diffusers/scripts/convert_original_stable_diffusion_to_diffusers.py --from_safetensors \ diff --git a/assets/20230708204347_1172778945_0_0.png b/assets/20230708204347_1172778945_0_0.png index 7d0fe7f..7dba320 100644 Binary files a/assets/20230708204347_1172778945_0_0.png and b/assets/20230708204347_1172778945_0_0.png differ diff --git a/assets/20230708204347_1172778945_0_2.png b/assets/20230708204347_1172778945_0_2.png index 2460691..5d08803 100644 Binary files a/assets/20230708204347_1172778945_0_2.png and b/assets/20230708204347_1172778945_0_2.png differ diff --git a/assets/20230902_tile_imgs.png b/assets/20230902_tile_imgs.png index 7e946ba..a247769 100644 Binary files a/assets/20230902_tile_imgs.png and b/assets/20230902_tile_imgs.png differ