Instructions to use Toc/toc with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Toc/toc with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Toc/toc", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| /* global titles:true */ | |
| // Mouseover tooltips for various UI elements. | |
| // `titles` is already defined by A1111, so we just merge into it... | |
| titles = { | |
| ...titles, | |
| "Dynamic Prompts enabled": "Disable dynamic prompts by unchecking this box.", | |
| "Combinatorial generation": ` | |
| Instead of generating random prompts from a template, combinatorial generation produces every possible prompt from the given string. | |
| The prompt 'I {love|hate} {New York|Chicago} in {June|July|August}' will produce 12 variants in total. | |
| The value of the 'Seed' field is only used for the first image. To change this, look for 'Fixed seed' in the 'Advanced options' section.`.trim(), | |
| "Max generations (0 = all combinations - the batch count value is ignored)": ` | |
| Limit the maximum number of prompts generated. 0 (default) will generate all images. Useful to prevent an unexpected combinatorial explosion. | |
| `.trim(), | |
| "Combinatorial batches": `Re-run your combinatorial batch this many times with a different seed each time.`, | |
| "Magic prompt": ` | |
| Magic Prompt adds interesting modifiers to your prompt for a little bit of extra spice. | |
| The first time you use it, the MagicPrompt model is downloaded so be patient. | |
| If you're running low on VRAM, you might get a CUDA error.`.trim(), | |
| "Max magic prompt length": | |
| "Controls the maximum length in tokens of the generated prompt.", | |
| "Magic prompt creativity": | |
| "Adjusts the generated prompt. You will need to experiment with this setting.", | |
| "Magic Prompt batch size": | |
| "The number of prompts to generate per batch. Increasing this can speed up prompt generation at the expense of slightly increased VRAM usage.", | |
| "I'm feeling lucky": ` | |
| Uses the lexica.art API to create random prompts. | |
| The prompt in the main prompt box is used as a search string. | |
| Leaving the prompt box blank returns a list of completely randomly chosen prompts. | |
| Try it out, it can be quite fun. | |
| `.trim(), | |
| "Attention grabber": `Randomly selects a keyword from the prompt and adds emphasis to it. Try this with Fixed Seed enabled.`, | |
| "Write prompts to file": ` | |
| The generated file is a slugified version of the prompt and can be found in the same directory as the generated images. | |
| E.g. in ./outputs/txt2img-images/.`.trim(), | |
| "Don't generate images": | |
| "Be sure to check the 'Write prompts to file' checkbox if you don't want to lose the generated prompts. Note, one image is still generated.", | |
| "Enable Jinja2 templates": | |
| "Jinja2 templates are an expressive alternative to the standard syntax. See the Help section below for instructions.", | |
| "Unlink seed from prompt": | |
| "Check this if you want to generate random prompts, even if your seed is fixed", | |
| "Don't apply to negative prompts": | |
| "Don't use prompt magic on negative prompts.", | |
| "Fixed seed": ` | |
| Select this if you want to use the same seed for every generated image. | |
| This is useful if you want to test prompt variations while using the same seed. | |
| If there are no wildcards then all the images will be identical. | |
| `.trim(), | |
| "Write raw prompt to image": | |
| "Write the prompt template into the image metadata", | |
| }; | |