polisteps

6 ms·

1.
by polisteps 14d ago
tracking all jev reproduction efforts
2.
Jev Reproductions Tracker(huggingface.co)
1 points by polisteps 14d ago|1 comments
3.
ZeroLabs – 100x cheaper than ElevenLabs (free forever locally) with open models(huggingface.co)
4 points by polisteps 3mo ago|1 comments
4.
by polisteps 3mo ago
While LLM and agents dominate the discourse, folks didn't realize that open Source has either caught u, approached or surpassed quality in voice tasks (text-to-speech, transcription, sound effect generation, etc). This demo running on
5.
Goodbye cold boot – how we made LoRA Inference 300% faster(huggingface.co)
1 points by polisteps 3y ago|0 comments
6.
by polisteps 3y ago
Two random LoRAs are loaded into SDXL, can you find a way to combine them?
7.
LoRA the Explorer(huggingface.co)
3 points by polisteps 3y ago|0 comments
8.
“Control Stable Diffusion” Sprint kicks off with free TPU-v4 from Google(github.com)
1 points by polisteps 4y ago|0 comments
9.
Train your own ControlNet models with diffusers(huggingface.co)
2 points by polisteps 4y ago|0 comments
10.
by polisteps 4y ago
GitHub: https://github.com/kakaobrain/karlo diffusers lib integration: https://github.com/huggingface/diffusers/releases/tag/v0.11....
11.
Karlo, the first open source DALL-E 2 replication is here(huggingface.co)
4 points by polisteps 4y ago|1 comments
12.
Dreambooth training UI for training a model for less than US$0.80(huggingface.co)
2 points by polisteps 4y ago|1 comments
13.
by polisteps 4y ago
Probably the cheapest UI, by hacking a bit Hugging Face Spaces. Also works locally. Of course Google Colab is still free but I think the UI / pre-curation of hyperparameters may be worth it?
14.
Diffusers 0.7.0: Stable Diffusion up to 2x faster on any hardware (including M1)(github.com)
1 points by polisteps 4y ago|1 comments
15.
by polisteps 4y ago
Pipelines also load faster, more schedulers are out, and the community contributed a bunch!
16.
Images in 8s on Colab Free Running Stable Diffusion on TPU(colab.research.google.com)
2 points by polisteps 4y ago|0 comments
17.
by polisteps 4y ago
Which is fairer imo than paying for the whole site. I prefer creator control on which content is paid and which content is not.
18.
Stable Diffusion pre-loaded with 250 community textual inversion concepts(huggingface.co)
4 points by polisteps 4y ago|0 comments
19.
by polisteps 4y ago
Hi, I'm the creator of multimodal.art, I didn't overlook it, but there's no "specialized" NSFW content maker to be highlighted - this Vice articles just show people using the model in different iterations to generat
20.
by polisteps 4y ago
Hi everyone! I am the creator of this website multimodal.art. Our goal is to break the barries of entry for this tech, as well as to inform people about the potentil of that technology. We also developed MindsEye, an open source pilot to ma
21.
Nvidia releases Alias-Free GAN code and pre-trained models, naming it StyleGAN3(github.com)
243 points by polisteps 5y ago|60 comments
22.
by polisteps 6y ago
Who else read the title and though it would render photo realistic glasses as filters in your face? :-)