ᴀʟᴠᴀʀᴏ

@Akira_VLRT same here :(

Guys, I just finished a comp game, closed my game and now I can't open Riot Client, screenshots below on what my issue is, 2nd one appears a bit after 1st one, can anyone help me? In-game for others it also appears as me being online.

@matchaman11 @Robertooai But what's the actual prompt for the sheet?

@Robertooai It's using Nano banana pro from github.com/Anil-matcha/Se…

ComfyUI workflow for consistent characters in Seedance 2.0 is finally cracked Train once → generate infinitely Breaking down the exact workflow in the thread 👇

@max__drake that's crazy, please explain a simple way to do this.

update: this worked????

@Stanny_Boiii @brave It's a big mess, especially for people with over 100+ tabs :(

@brave W, fully switched back to brave cause of this.just need to sort out your vertical tabs

Brave now has an ultra dark theme! 🌑 To use it, first make sure you've updated Brave for desktop to v1.88. Then, go to Settings -> Appearance -> Theme and toggle "Ultra dark theme" at the bottom.

@lookner 7-8 stratotankers underway to iran right now!!

@ostrisai @tawnkramer Can't they "license" their likeness in order for someone to use/train it?

@tawnkramer The issue is you would own the rights to the images, not their likeness. For instance, you couldn't train a video model on the images and make a movie with it. Or make an AI content creator based on their likeness from the images. Open source likeness would allow for all of that

I really wish some people would open source their likeness. Put out a huge open "package" of themselves. Video, images, audio, etc. Free to use. Would be great for AI researchers to use for character training testing and they would become semi famous in the process. Who is in?

@Machinedelusion wow!

Got Training working inside comfy for AceStep 1.5! Doing a longer training run now to triple check my implementation is correct. If i get the green light i'll be releasing the node pack soon! Stay Tuned.

@acemusicAI This is amazing! great job :D

We're releasing ACE-Step-v1.5(2B), a fast, high-quality open-source music model. It runs locally on a consumer-grade GPU, generates a full song in under 2 seconds(on an A100), supports LoRA fine-tuning, and beats SUNO on common eval metrics. GitHub: github.com/ace-step/ACE-S… Key traits: Quality: beats Suno on common eval scores Speed: full song under 2s on A100 Local: ~4GB VRAM, under 10s on RTX 3090 LoRA: train your own style with a few songs License: MIT, free for commercial use Data: fully authorized plus synthetic The music AI space lacks commercial-grade open models. Many creators are forced to rely on closed-source services, and can’t fully own, run locally, or fine-tune their own models. We want to help change that.

@ReflctWillie exactly, that's why we need behind the screen presets.

Have been digging deeper into the AI influencer studio from Higgsfield. If you're building AI tools you should take note here. It's so... easy. Bing bang boom 'influencer'. AI has an accessibility issue, not in a WCAG way, in a 'IDFK what to prompt' way. The character creation UI dropping right into Kling motion control... the prompt field is borderline hidden. This is all visual, I didn't type a single character. Normal people don't know how to prompt. If you're making generative AI tools for normal people to use - start with the training wheels before you show them the BMX track.

When you're working with big boy prompts, you need big boy readability. - Improved editing experience for prompt nodes. Lots of room for more improvement here. - Better UX for running image gen nodes. Shout out @djalvaro for that one. User feedback, hmu.

@cocktailpeanut @Bazed_X_ Just need to be able to train as well and better quality models

Yup -- i don't think people yet realize how game changing this is. The model has existed for a while but without a good user interface -- @Bazed_X_ has made it accessible to all with SongGeneration Studio. Infinitely Free. No censorship. No watermarks. Just your PC.

@cocktailpeanut @Bazed_X_ The future for sure!

@ReflctWillie amazing update!

@ReflctWillie yeah feels a bit more intuitive to have it on the node and maybe the main run to just run only the entire thing if needed. Cant wait!

@djalvaro In the run button there is an option for 'run selected node only' and there is regenerate button on the node. But the ux is a bit clumsy. I'll implement something like what you're describing in the next release.

Node Banana must grow. - Discord server: link below, hmu. - Generate node carousel: the active image will be used in the workflow. - Node group locking: because I accidentally hit run and it killed the mood. - Hot hot keys: Shift + (p)rompt, (g)enerate image, (i)mage, (l)lm node. Also cmd + enter to run. - Logging - Split grid node can now be run manually.

@Yatharth3501 Awesome, is there a way to improve 44.100 songs or not really made for that?

I just officially released FlashSR, a model that can massively enhance audio quality at speeds over 200x realtime. This was used in MiraTTS, but now anyone can use it for their own models. Repo: github.com/ysharma3501/Fl… Model: huggingface.co/YatharthS/Flas…

@ComfyUI This is great!

We’re exploring a new "Simple Mode" It’s intended to make complex workflows easier to share and iterate on, with the focus on results rather than the underlying node graph, especially for users who find large graphs overwhelming. We’d love to hear your thoughts. How would you use this feature?

@FlippingSigmas @Alibaba_Wan Nice! We could really use some frame options between First and End frames.

Dynamic Keyframing Node for @Alibaba_Wan Vace inbetweening - dynamically adjust where you want your keyframes in a sequence with visual sliders.

@ReflctWillie oh and more grid cutters for different aspect ratio's XD

@djalvaro Covered in detail in this post: x.com/ReflctWillie/s…

Quality of life updates just released for Node Banana - image grid splitting: should work with any grid layout - resizing improvements when you have multiple nodes selected - image history, so you don't lose images when regenerating.

@YouAreTheChaff @JoePro @minchoi Sure, but do you really believe gen-AI is subject to regulations? when at some point the difference between human output and AI output is going to be 0. And each and every future product will have some % of gen-ai.

It's happening Google removes AI videos of Disney characters after cease and desist letter 👀