Generative AI specialist /Motion Graph Working on AI first projects, XR games, documentaries and films. ComfyUI, After Effects, Ableton, Claude Code.
Finland
Joined March 2016
- Tweets9.1K
- Following695
- Followers1K
- Likes8.9K
Heba AI retweeted
We're giving away a Hogwarts Legacy custom wrapped NVIDIA GeForce RTX 5090 to celebrate its official release with DLSS 5 🪄
To enter:
🟢 Share this post
🟢 Comment #RTXON
Terms and Conditions: nvidia.com/en-us/geforce/con…
More testing with H3 model. Now vs Qwen 2.1
H3 left - Qwen right
H3 needs to be used with the 3904×2176 resolution or it will mess up the small faces. Qwen can use much smaller res and no problems with smaller faces. H3 is also quite plastic by default.
This bug has been quite major token eater.
Community note on this is more interesting than the article itself.
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
H3 edit image model is really good!
Today, we open-source P-H3-Image, an image generation and editing model built from MiniMax-H3 by @MiniMax_AI. A SOTA video model, turned into an efficient open image model.
- 𝗛𝗼𝘄 𝗲𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝗶𝘀 𝗶𝘁? Generate or edit images 5× faster, in 8 steps instead of 50. A 4 MP image takes 2.6 s instead of 12.4 s. The model is 7× smaller than H3, so high-resolution images are a light download and run on a much smaller GPU.
- 𝗛𝗼𝘄 𝗴𝗼𝗼𝗱 𝗶𝘀 𝗶𝘁? Strongest at photorealism, documentary and editorial photography, film stills, and graphic styles with short quoted text. It renders 2 to 16 MP and edits with up to 8 input images.
-𝗛𝗼𝘄 𝗱𝗼𝗲𝘀 𝗶𝘁 𝘄𝗼𝗿𝗸? We reworked every component of H3: the prompt upsampler, the text encoder, the diffusion schedule, the transformer, and the decoder. Use it with Diffusers or ComfyUI.
📄 Tech Report: pruna.ai/blog/oss-p-h3-image…
🤗 @huggingface Open Weights: huggingface.co/PrunaAI/P-H3-…
👉 @ComfyUI Native Support: huggingface.co/PrunaAI/P-H3-…
I forked this for 32gb vrams.
I deleted all Linux, Mac and DGX codes and cleaned 40,000 code lines.
I have made comfyui apps for years, so this will have faster gens than original repo. If somebody is interested I can upload it to Github.
@ComfyUI #OpenSourceHiggsfield Submission:
Instead of 1 feature, we built the entire Higgsfield platform + way more... but open source.
🎬 Imagination Studio.
An all in one studio with 61+ apps. Video,Music,Audio, Sfx/Foley, Agents, Prompt Assistance, Depth Mapping, Audio Separation, Loras, Custom 3D animatics, Editing.
Collab by Me and @u1tra_instinct
REPO LINK BELOW: 👇
github.com/drowzeys/Imaginat…
It's Monday morning and time to try something new: @t3dotcodes
I'm happy with my current CLI workflows, but hustling with various subs is a drag and I do dig how Theo rolls.
I did try T3 Code v 0.0.28 but it was meh, but now its in v.0.1.6 so time to give another chance.
Heba AI retweeted
Defined roles and “agent” archetypes are so 2024 lol
We are definitely exploring “bot” like systems, specifically meta threads for commanding your threads/agents/envs, but it will not be a defined role system because those are artifacts of a time where models were much dumber.
Hacker Tube, like YouTube but better.
Comes with Longs instead of Shorts.
Weekend hack: HackerTube. I was annoyed with YouTube's recommendations so I had Claude filter down 43M comments over 20 years of HN to 6,454 videos that keep being recommended. *Significantly* better viewing experience, ~2 hours with Opus 5.5:
hackertube.org/
- Excellent recs without any logins (state is kept in local browser session)
- Thumbnail is a representative frame from middle of the video instead of the super duper A/B test hyper-optimized user uploads
- Underlying data from the comment analysis as one zip file, point your Claude at it and ask for your own version
- HackerTube Longs ✊
A bit more on the process:
I use Claude Code (desktop app for Mac), Opus 5.5 on Medium effort as my default. I only talk to Claude and have it do everything, the only "manual work" I helped with in tandem was in the browser (reminiscent of MenuGen): 1) I had to create some kind of API key from Google Cloud UI to query YouTube for video metadata, 2) I booted up a VPS box for serving and 3) I bought the domain name. Once the box was up I gave Claude the IP and told it update, secure, harden, deploy and optimize the code, and it did the whole thing.
Many awesome projects that would have normally taken weeks/months even can now be few-shotted for breakfast. What else are we not thinking of.
Heba AI retweeted
@higgsfield_ai sent us a cease and desist last night, asking us to shut down our Open Source Higgsfield challenge.
Our read on why they sent it. When the apps people pay for can be rebuilt in the open in two weeks, that’s an uncomfortable thing for a closed platform to watch.
So here’s us reiterating our promise to every builder in the challenge and community, and we’re keeping it:
• The challenge stays open
• Final submissions close tomorrow, Oct 30
• Winners are announced Nov 6, with prizes paid as promised
You showed up for each other. We’re showing up for you.
If we win, we all win!
Comfy out 🫡
Last night, @higgsfield sent @ComfyUI a cease and desist (attached)
They want the Open Source Higgsfield challenge taken down within 48 hours.
Their argument is that our use of the Higgsfield name and that using their CEO's likeness is an infringement violation (man, the irony). Claiming that it makes the event look like Higgsfield is associated with it.
This couldn't be farther from the truth, which is that Higgsfield made a promise to open source its product and never did.
For the past two weeks, builders have been recreating popular closed apps as open workflows anyone can run.
So we asked our community to deliver where Higgsfield failed: apps in the open, built by the community, for the community. We've made it clear that the challenge is organized by ComfyUI and not affiliated with or endorsed by Higgsfield.
When we started Comfy, we made the community a promise. Our product stays open, free, and yours. We’ve kept it every day since, and one letter isn’t going to change that.
The challenge stays up. Comfy and open source will win!
