Jennifer Connelly is an American actress, known for many roles in films such as Labyrinth, Career Opportunities, Hulk and Requiem for a Dream, among many others. She won an Oscar for Best Supporting Actress for her role in A Beautiful Mind (2001).
Appreciate my work? My TIs are free, but you can always buy me a coffee. :)
Curious about my work process? I have summarized it here.
Updated! Please check the "About this version" section for more info.
How to create a good prompt using my TIs
You're obviously free to experiment, but bear in mind that my TIs are trained with a more or less fixed phrasing, that normally starts with:
"photo of EMBEDDING_NAME, a woman"
So I recommend always starting your prompt like that and then building the rest of the prompt from there. For instance, "photo of beautiful (jg_jc0n:0.99), a woman, (sfw tight fashion shirt:1.2), (long jeans), in a departmentn store, natural light, (masterpiece:1.2) (photorealistic:1.2) (best quality) (detailed skin:1.2) (intricate details) (8k) (HDR) (cinematic lighting) (sharp focus), (looking at the camera:1.1), (closeup portrait:1.1), (painted lips)"
Description
Retrained from scratch with a new dataset. This time I managed to get ahold of a few decent quality pics, and I think that shows in the final result. The embedding I'm uploading is step 1850. I trained at a 0.004 learning rate until step 1000. From there to step 2500, at a 0.001 rate. There were signs of saturation from very early on the training process, but by simply reducing the embedding weight in the prompt to 0.99 you will get rid of most of said saturation.
FAQ
Comments (24)
how to use textual inversion please help
You just download it, place it in the embeddings folder (stable-diffusion-webui\embeddings), then use the embedding's name in your prompt.
Unless you are on Easy Diffusion (like me), then just forget about it... C'mon, ED team, bring textual inversion.
Hey can you add Alexa Rivera to the list
Famous youtuber with 12.4 M subscribers
https://**www.mediafire.com/file/43jh0aie0roxmhe/LexiRivera.zip/file
Added! Beautiful girl. I'll probably start working on her TI soon. :)
I prefer Jennifer Lynn Connelly from the time period of 1985's Phenomena to 1990's The Hot Spot. It was her angelic and beautiful era. After 1991, it should belong to her devilishly sexy era.
Great stuff.
Early Jennifer Connelly was a looker for sure.
If you're open to training a LoRA I'll be back with another review!
Thank you for the comment! :) I wish I knew how to traing a LoRA... never tried yet! xD For the time being, I'll keep sticking to TIs, as I still have a few in mind (and a few requests too), but in the future I may try to at least get a taste of what training a LoRA is. :)
@JernauGurgeh believe me pal, if you can make a TI of that quality, making a LoRA shouldn't be a problem
Either way, please proceed at your leisure, I'll keep myself updated on your releases
Take care!
@EDG Thanks, pal! Your collection of LoRAs (particularly the Haute Couture ones) has been a favorite of mine for quite some time now, so this coming from you is a great compliment. :)
Amazing work,im really big fan of her and in general everything from 80s and early 90s.
Thanks to AI she can live at her prime again.
Keep up with good work!
Thank you for the comment! :)
@JernauGurgeh Yeah absolutely,i render her with my custom model that is built for photorealistic humans,and she looks gorgeous
Real blast from the past!
@grimreaperalphax007gmailcom814 Really glad to hear that. This was definitely one of the embeddings I always had in mind before I started doing TI training. One of my favorite actresses of all time and, during that era, probably one of the prettiest and hottest women in the world.
guys, is pickle tensor safe like safetensors? I've only using safetensors so far
I'm not sure what the difference between the two formats actually is, but, if it's any help, you can totally trust me. ^^ I created this TI myself and I know basically nothing about programming, so I guess I wouldn't be able to add something malicious to my embeddings even if I wanted to. :D
@JernauGurgeh ok bro thanks for the honesty, thanks for the work I'll try using it :)
No, pickle tensor can potentially contain arbitrary code. There are some things you can do to check that improve the safety, here is a video explaining: https://youtu.be/Qto8R9j6VaI
(Not my video, just linking it to be helpful)
@JernauGurgeh a pickle is capable of running code on anyones machine so they're avoided.
@borkborkbork Damn, I missed @kajuberdut's comment. I'll have a look at that video and see if I can do something about it. Thanks for the heads-up, both of you!
What happened to all the sexy Bikini pictures?!
LOL . . . I know, I know... new Civitai rules !
Version 1 was really good, but I will try the new version 2.1 . Thank You JearnauG
I'll be honest, it really hurts to delete some of these pics but... it is what it is. xD Thanks for your support, man!
Recently I got also a lot of removals from my models, all those images with bikini, cleavage or any kind of sexual exposures were marked as an infraction. So what it's left now are those "safe" pictures.
@smoonfucker you are not alone. Some of my models also got deleted or the images banned.
So I erased all my models out of here. Its not worth it.
I post them in other places where people respect the Artist´s Work.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.