90 detailed dynamic prompts with fantasy, sci-fi, video game, mythological and historical themes. Each prompt includes setting, background, pose and (mostly) gender-agnostic character details. Now with booru tags!
For ComfyUI, please use https://github.com/adieyal/comfyui-dynamicprompts, and https://github.com/GrvBdgr/comfyui-negativewildcardsprocessor.
The wildcards contain negative wildcards (text surrounded by <! .. !>). If you do not wish to use these, simply remove the <! .. !> text in the wildcards files with your favorite text editor. Use the following regex:
<!.*!>As of v3.0 the wildcards no longer require Loras or embeddings, for compatibility with all models.
Older versions of the wildcards are not recommended for use.
Usage:
Add a description of the character you'd like in the generated image (feel free to use character LoRas), plus any style/quality keywords necessary for the model you're using, then add __ffpc-booru__.
Example:
best quality, masterpiece, 1man, russian (male:1.1), solo, aquamarine eyes, long brown beard, __ffpc-booru__
Some example characters are provided in the character.txt in the downloaded archive, so if you just want to run a few tests, you can use the following prompt:
best quality, masterpiece, __character__ __full-prompt-fantasy__
More diverse character wildcards can be found here: https://civarchive.com/models/118508?modelVersionId=128565
Negative prompt should be whatever the model necessitates, or your preferred negative prompt. Some models have race or gender bias due to the material that was used for training, you can usually use negative prompts to compensate. eg. if your generations come out with female characters despite describing the character as male, add "woman, girl" to the negative prompt, use weights if necessary.
Additional usage notes:
Other than the above, feel free to experiment with additions to the prompt, though most of the wildcards are rather long, so if you want to have a significant impact, weights may be necessary. The prompts rarely use weights above 1.1, so if your additions seem to have no effect, try weights of 1.15 or more.
Description
LoRa & Embedding downloader script added (see description)
New LoRas added, more negative prompts added, overall improvement of old wildcards.
FAQ
Comments (21)
I want to tell you how amazing this is. Who needs to think of a prompt when this tool!
This happen when I try to run the lora_downloader .bat file
import requests
ModuleNotFoundError: No module named 'requests'
D:\_Stable Diffusion\stable-diffusion-webui\extensions\Umi-AI-Embeds\wildcards\fullFeatureFantasy_v21>pause
Press any key to continue . . .
Your python installation doesn't have the requests package. Open a command prompt and execute the following command:
pip install requests
or if that fails, try the following:
python -m pip install requests
Once it finishes, try running the batch file again.
Also, judging by the folder structure in your error message, you've created an extra folder within your wildcards directory ("fullFeatureFantasy_v21"). The downloader script currently only works properly, if it is directly under the wildcards folder, as it looks for the embeddings and Lora folders at a relative path to itself. In the next version, I'll try to come up with a solution that allows for nesting the script deeper into the wildcards folder.
I removed the folder and it start downloading,however full of error msg like this
Error downloading file (Attempt 4/5):
Retrying in 10 seconds...
Downloading https://civitai.com/api/download/models/23343 (Fire_VFX-000010.safetensors)...
Error downloading file (Attempt 5/5):
Max retry attempts reached. Download failed for https://civitai.com/api/download/models/23343 (Fire_VFX-000010.safetensors).
Downloading https://civitai.com/api/download/models/12758 (waterVFX.safetensors)...
Download completed: D:\_Stable Diffusion\stable-diffusion-webui/models/Lora\waterVFX.safetensors
Downloading https://civitai.com/api/download/models/37208 (bronzepunkai.safetensors)...
Error downloading file (Attempt 1/5):
Retrying in 10 seconds...
@mightyckf858 If some of the links are downloading fine, while others result in timeout, it might be due to a couple of reasons.
- check if you can download the problematic link manually, if not, then report it to me, as the link may be broken
- make sure that you don't have anything else running in the background, that may heavily influence internet traffic (like a bit torrent client)
- it may just be the case, that civitai servers were busy during the time you tried, so it may help if you repeat the download at a different time of day
where should the embendings and loras from your download script go? i have my loras and embedings on a external SSD.
The script was designed with the A1111 folder structure in mind.
You've got your wildcards (and the script) here:
...\stable-diffusion-webui\extensions\<extension_name>\wildcards
The script will assume that your embeddings and LoRas are under the following paths, and will try to download them there:
...\stable-diffusion-webui\embeddings
...\stable-diffusion-webui\models\Lora
I haven't considered alternative folder structures, though it's a fair point. In the next version I'll add a fallback, so that in case these folders are not found, the respective models will just be downloaded into a new folder next to the script.
@TxcTrtl thanks that helps
@Jesse_F I've been using junctions (not symbolic links) to make sure everything can see my models and embeddings. I heartily recommend using Link Shell Extension (LSE) if you intend to use this method. I can help you set that up if you need.
Nice wildcards, what is <lora:ldv2:0.6> link, I cant find it.
thx, you change lots of lora's name,
@xueqing12211 Quite the contrary, I tend to keep the name that the downloaded files have. Names shouldn't be a problem though, if you use the included downloader script to get all the LoRas
Awesome update, thanks for creating the download script!
While looking through the code, I couldn't tell how it handles some of the loras/embeddings already being there and didn't want to risk it re-downloading everything so I added this after line 44:
if os.path.exists(file_path):
print(f"File {file_name} already exists. Skipping download.")
return
Looks like the formatting got messed up. Print and return should of course be indented.
@theunlikely Nice catch. I'll add it in the next version.
Unless im mistaken the link for MagicalCircle_Sora is invalid. which Lora is this?
Original Link: https://civitai.com/api/download/models/123171
I'm afraid the model is no longer available on civitai. It used to be here: https://civitai.com/models/105512. I'll remove it from the wildcards in the next iteration.
Once again. One of the best things of this site. Congrats!
File 'path/to/the/links/file\fpf_download_links.txt' Not Found
It's odd, I've had mix path seperators work just fine in other scripts. (Note the change from "/" to "\")
The lora_downloader.py script uses "/" pretty much everywhere else, but for some reason changes to "\\" (escaped "\") in the line:
>> download_links = read_file(script_directory + "\\" + LINKS_FILE)
If anyone else has trouble with the links 'File Not Found', change the "\\" to "/", as in:
>> download_links = read_file(script_directory + "/" + LINKS_FILE)
Or, more universal path joining:
>> download_links = read_file(os.path.join(script_directory + LINKS_FILE))
Though, the other dir references such as ... + "/embeddings", and ... + "/models/Lora" would need to be wrapped in os.path.join() to make the script fully universal.
Again, I've used mixed path separators in other scripts with no problems. Maybe it's something specific to the venv I was in?
Anyhow, if anyone else gets a 'file not found' for the links file, even though it is actually there in the right place, this should get you fixed.
Thanks for the suggestion, corrected this in v2.2.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.
















