Errthang is alive retweeted
There's a buzzy new paper about GPT-6 Astra being able to directly interpret model weights (prompts full of floating point numbers) - I tried it with Qwen 3.8 27B on my laptop and found that, after 45+ minutes, that model could achieve the same thing simonwillison.net/2026/Nov/3…
Errthang is alive retweeted
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-…
Errthang is alive retweeted
Drag-and-drop makes the digital graspable, but building it well is fiddly enough that many don’t bother. Declarative Dragging (with Josh Horowitz & Jeff Heer, #uist2026 best paper award), lets you make an element draggable just by listing the states dragging it can reach.
Errthang is alive retweeted
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.…
In 2019, Kate Crawford and Trevor Paglen trained a model only on the “person” categories of ImageNet. They called it ImageNet Roulette. A user uploaded a photo, and the model gave the person a label.
excavating.ai/
Errthang is alive retweeted
Levitated lives again! Thanks to Ruffle, my Flash work from two decades ago is live and alive!
levitated.net/?v=2
Same form, /w flowers instead of branches: calla lilies for Shin and Soe, tulips for Hikae, in a black dish. From the front it's easy to forget the back. Take away the stems behind Shin and Soe and the work goes flat. Put them in, it gains depth and quiet support. #cavalryapp
Errthang is alive retweeted
We released the code and model weights here: github.com/ayushjain1144/tra…. Have fun! 🙂
1/ We introduce TrackEverything: a 3D point tracker that tracks all points across all frames of long videos (1000+ frames).
Our key idea is to tie computation to unique 3D scene content rather than redundant 2D pixels in a video.
trackeverything.github.io
The Origin Form- Every Sogetsu arrangement starts here.
Here it's raspberry branches for Shin and Soe, with rapeseed for Hikae. The detail that makes or breaks it is the gap between Soe and Hikae. Too wide and the form falls apart; too narrow and it feels cramped.
#cavalryapp