--- library_name: diffusers pipeline_tag: text-to-image datasets: - hearmeneigh/e621-rising-v2-curated - hearmeneigh/e621-rising-v1-curated tags: - anthro - furry - e621 - nsfw - not-for-all-audiences ---

Outdated!

This model has been superseded by:

E621 Rising V3 (SDXL)

Warning: This model is NOT suitable for use by minors. The model can/will generate X-rated/NFSW content.
Ckpt | Safetensors | Yaml | Tags | Example Prompts | WebUI How-To
# E621 Rising V2: A Stable Diffusion 2.1 Model [epoch 29] * Guaranteed **NSFW** or your money back * Fine-tuned from [Stable Diffusion v2-1-base](https://huggingface.co/stabilityai/stable-diffusion-2-1-base) * Training continued from [E621 Rising V1](https://huggingface.co/hearmeneigh/sd21-e621-rising-v1) * 10 additional epochs of 250,000 images each, collected from [E621](https://e621.net/) * Trained with [6,246 tags](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/blob/main/meta/tag-counts-by-name.json) * `512x512px` * Compatible with 🤗 `diffusers` * Compatible with `stable-diffusion-webui` * Compatible with anything that accepts [`.ckpt`](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/resolve/main/sd21-e621-rising-v2-epoch-29.fp16.ckpt) and [`.yaml`](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/resolve/main/sd21-e621-rising-v2-epoch-29.fp16.yaml) files ## Getting Started * [Stable Diffusion WebUI How-To](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/blob/main/guides/WEBUI.md) * [Python How-To](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/blob/main/guides/PYTHON.md) * [Prompt Guide](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/blob/main/PROMPT-GUIDE.md) ## Examples ![Richie3](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/resolve/main/examples/richie3.png) [More examples and prompts here](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/blob/main/PROMPT-GUIDE.md#prompt-examples) ## Versions | Precision | CKPT | Safetensors | YAML | Notes | | :-------- | :--- | :---------- | :--- | :---- | | `FP16` | [Download](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/resolve/main/sd21-e621-rising-v2-epoch-29.fp16.ckpt) | [Download](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/resolve/main/sd21-e621-rising-v2-epoch-29.fp16.safetensors) | [Download](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/resolve/main/sd21-e621-rising-v2-epoch-29.fp16.yaml) | Use this by default | | `FP32` | [Download](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/resolve/main/sd21-e621-rising-v2-epoch-29.fp32.ckpt) | [Download](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/resolve/main/sd21-e621-rising-v2-epoch-29.fp32.safetensors) | [Download](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/resolve/main/sd21-e621-rising-v2-epoch-29.fp32.yaml) | | | `BF16` | [Download](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/resolve/main/sd21-e621-rising-v2-epoch-29.bf16.ckpt) | [Download](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/resolve/main/sd21-e621-rising-v2-epoch-29.bf16.safetensors) | [Download](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/resolve/main/sd21-e621-rising-v2-epoch-29.bf16.yaml) | | ## Changes From E621 See a [complete list of tags here](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/blob/main/meta/tag-counts-by-name.json). * Symbols have been prefixed with `symbol:`, e.g. `symbol:<3` * All categories except `general` have been prefixed with the category name, e.g. `copyright:somename`. The categories are: * `artist` * `copyright` * `character` * `species` * `invalid` * `meta` * `lore` * Tag names are all lowercase and only contain `a-z`, `0-9`, `/`, and `_` letters * `:` is used to separate the category name from the tag ### Additional Tags * Image rating * `rating:explicit` * `rating:questionable` * `rating:safe` ## Training Procedure [Training legend](https://huggingface.co/hearmeneigh/sd21-e621-rising-v2/blob/main/examples/training-legend.jpg) * 204-272 images per batch (epoch variant) * `512x512px` image size * Adam optimizer * Beta1 = `0.9` * Beta2 = `0.999` * Weight decay = `1e-2` * Epsilon = `1e-08` * Constant learning rate `4e-6` * `bf16` mixed precision * 8 epochs of [V1 dataset](https://huggingface.co/datasets/hearmeneigh/e621-rising-v1-curated) samples stretched to `512x512px` (ignore aspect ratio) * 9 epochs of V1 dataset samples resized to `512xH` or `Wx512px` with center crop (maintain aspect ratio) * 2 epochs of V1 dataset samples resized to `< 512x512px` (maintain aspect ratio) * 10 epochs of [V2 dataset](https://huggingface.co/datasets/hearmeneigh/e621-rising-v2-curated) samples resized to `< 512x512px` (maintain aspect ratio) * Tags for each sample are shuffled for each epoch, starting from epoch 16