jbilcke-hf (Julian Bilcke)

posted an update 29 days ago

Post

3212

Did you know that there is a UI wrapper around https://github.com/a-r-r-o-w/finetrainers which is a great library made by @a-r-r-o-w for finetuning AI video models?

The UI is called VideoModelStudio (or VMS in casual chat)

All you have to do is to duplicate this space:
jbilcke-hf/VideoModelStudio

2 replies

·

posted an update 29 days ago

Post

1865

Hi everyone,

I've seen some unsuccessful attempts at running Wan2GP inside a Hugging Face Space, which is a shame as it is a great Gradio app!

So here is a fork that you can use, with some instructions on how to do this:

jbilcke-hf/Wan2GP_you_must_clone_this_space_to_use_it#1

Note : some things like persistent models/storage/custom LoRAs might not be fully working out of the box. If you need those, you might have to dig into the Wan2GP codebase, see how to tweak the storage folder. Happy hacking!

reacted to merve's post with 🤗🔥 5 months ago

Post

5457

Oof, what a week! 🥵 So many things have happened, let's recap! merve/jan-24-releases-6793d610774073328eac67a9

Multimodal 💬
- We have released SmolVLM -- tiniest VLMs that come in 256M and 500M, with it's retrieval models ColSmol for multimodal RAG 💗
- UI-TARS are new models by ByteDance to unlock agentic GUI control 🤯 in 2B, 7B and 72B
- Alibaba DAMO lab released VideoLlama3, new video LMs that come in 2B and 7B
- MiniMaxAI released Minimax-VL-01, where decoder is based on MiniMax-Text-01 456B MoE model with long context
- Dataset: Yale released a new benchmark called MMVU
- Dataset: CAIS released Humanity's Last Exam (HLE) a new challenging MM benchmark

LLMs 📖
- DeepSeek-R1 & DeepSeek-R1-Zero: gigantic 660B reasoning models by DeepSeek, and six distilled dense models, on par with o1 with MIT license! 🤯
- Qwen2.5-Math-PRM: new math models by Qwen in 7B and 72B
- NVIDIA released AceMath and AceInstruct, new family of models and their datasets (SFT and reward ones too!)

Audio 🗣️
- Llasa is a new speech synthesis model based on Llama that comes in 1B,3B, and 8B
- TangoFlux is a new audio generation model trained from scratch and aligned with CRPO

Image/Video/3D Generation ⏯️
- Flex.1-alpha is a new 8B pre-trained diffusion model by ostris similar to Flux
- tencent released Hunyuan3D-2, new 3D asset generation from images

7 replies

·

posted an update 7 months ago

Post

15427

Doing some testing with HunyuanVideo on the Hugging Face Inference Endpoints 🤗

prompt: "a Shiba Inu is acting as a DJ, he wears sunglasses and is mixing and scratching with vinyl discs at a Ibiza sunny sand beach party"

1280x720, 22 steps, 121 frames

There are still some things to iron out regarding speed and memory usage, right now it takes 20min on an A100 (see attached charts)

but you can check it out here:

https://huggingface.co/jbilcke-hf/HunyuanVideo-for-InferenceEndpoints

There are various things I want to try like the 100% diffusers version and other models (LTX-Video..)

posted an update 12 months ago

Post

32549

I am working to add automatic speech bubbles to the AI Comic Factory!

This is not finished yet - there are a few things to tweak - but so far results are pretty promising!

(it's available through a "beta" toggle)

9 replies

·

replied to their post about 1 year ago

Hello, it looks like it was a temporary issue maybe (I was in time off yesterday, but i've checked the comic factory today and it seems to work fine)

replied to their post about 1 year ago

Hello everyone,

this space is still pretty new and experimental, it might be paused or break sometimes - sorry about that!

I'm still actively developing (for instance I'm going to add a way to extend a video project)

Don't forget to check the Prompt Academy to learn a bit more about the project:

https://latent-store.notion.site/AI-Stories-Academy-8e3ce6ff2d5946ffadc94193619dd5cd

posted an update about 1 year ago

Post

30044

Added a new button to the AI Stories Factory for more fun!

jbilcke-hf/ai-stories-factory

11 replies

·

posted an update about 1 year ago

Post

7092

If you haven't done it yet,

try my new AI Stories Factory space it is pretty fun!
jbilcke-hf/ai-stories-factory

You can even prompt the visual and music style 💅

2 replies

·

reacted to nisten's post with 🔥 about 1 year ago

Post

5799

Just had the former chief public health officer of the Netherlands🇳🇱 review a huggingface AI-doctor I made via a simple orpo-zephyr-8x22b-GPT and they think it's really good.

https://hf.co/chat/assistant/661d77310e3aea9ae571e43c

3 replies

·

replied to their post about 1 year ago

Hi, it looks like it was a temporary issue, which repaired itself automatically

for some context the comic factory isn't built as solid as say a banking app, or a Netflix cloud monitored 24/24 by humans, it is more experimental in nature: the servers can sometime crash, stop, be re-allocated to other Hugging Face Spaces

Usually when that happen the servers and the comic factory will restart and work again automatically after 10-15 min

if a downtime lasts longer than this autostart period, you can ping me on X https://twitter.com/FLNGR and I will restart it manually 🤗

posted an update about 1 year ago

Post

6516

I'm extending the AI Comic Factory (to follow-up on the "bring your own model" philosophy 🤗) to support the broader LLM ecosystem of our other vendor friends!

Here is the announcement:
jbilcke-hf/ai-comic-factory#723

This is an experimental feature, some models/vendors might require parameter tuning but I haven't tested all of them yet (only a bit of GPT-4 👀)

Let me know if you experience any issues!

jbilcke-hf/ai-comic-factory

4 replies

·

posted an update over 1 year ago

Post

8184

The AI Comic Factory 1.2 has been released a few hours ago!

You can now tell the LLM in advance how many pages you will want, for richer and deeper stories

https://www.loom.com/share/b8b2d55fc60249a78df60adaa2673d2f

I the future I would like to improve how layouts are attributed to pages ( seeing the same layout 6x can be a bit boring)

Also, I'm still waiting for some new improved AI models to drop, so stay tuned!

2 replies

·

reacted to mrfakename's post with 🔥 over 1 year ago

Post

4266

Today, I'm excited to launch two new models on the TTS Arena: MeloTTS and StyleTTS 2. Both are open sourced, permissively licensed, and highly efficient.

Curious to see how they compare with other leading models? Vote on the TTS Arena ⬇️

TTS-AGI/TTS-Arena

MeloTTS, released by MyShell AI, provides realistic and lifelike text to speech while remaining efficient and fast, even when running on CPU. It supports a variety of languages, including but not limited to English, French, Chinese, and Japanese.

StyleTTS 2 is another fully open sourced text to speech framework. It's permissively licensed, highly-efficient, and supports voice cloning and longform narration. It also provides natural and lifelike speech.

Both are available now to try on the TTS Arena - vote to find which one is better! The leaderboard will be revealed once we collect enough votes.

14 replies

·

posted an update over 1 year ago

Post

Working on a new open-source project to showcase levihsu/OOTDiffusion 🤗

But it's not a Space 🫢 Can you guess what it does? 👀

reacted to mrm8488's post with 🤝👍🤗 over 1 year ago

Post

Hello world! 🔥

reacted to mariagrandury's post with ❤️ over 1 year ago

Post

Holiday talk about AI taking over? Let's shift the narrative!

🌟 There is no reason to believe that just because AI systems are intelligent they will want to dominate us. Yann LeCun reminds us that AI systems won't have the same motivations as humans, we'll design them not to.

🌍 Instead of getting distracted by future existential risks, we must address AI’s more pressing risks — like emitting carbon, infringing copyrights and spreading bias. Sasha Luccioni urges us to create tools and legislation that promote transparency and diversity.

💡 Dive deeper into these perspectives:
- Yann's ( @ylecun ) WIRED interview (12'): https://www.wired.com/story/artificial-intelligence-meta-yann-lecun-interview/
- Sasha's ( @sasha ) TED Talk (10'): https://www.ted.com/talks/sasha_luccioni_ai_is_dangerous_but_not_for_the_reasons_you_think

P.S.: Love this new "Posts" feature, big thanks to 🤗 for letting me try it!

What are your go-to citations for AI risks? 👇

2 replies

·

Julian Bilcke

AI & ML interests

Recent Activity

Organizations

Julian Bilcke

AI & ML interests

Recent Activity

Organizations

jbilcke-hf's activity