# Show HN: Weight-space programming playground with WebGPU

> Source: <https://tabbyml.github.io/weight-space-playground/>
> Published: 2026-11-19 15:32:47+00:00

**Key takeaway:** Editing just 0.016% of Qwen3.5-0.8B's weights, without gradient descent, reduced reasoning loops in the demo, and the WebGPU playground lets developers run the inspect-edit-evaluate loop in a browser.

I wanted to build a full weight-space programming loop entirely in the browser with WebGPU: let an LLM inspect Qwen3.5-0.8B, edit its weights, evaluate, and repeat.

the full loop needed too much VRAM. so I turned it into an interactive walkthrough of a concrete example: editing 0.016% of a model's weights to reduce reasoning loops, no gradient descent.

Based on exp8 of “Sparks of Weight-Space Programming.”

[https://tabbyml.github.io/weight-space-playground/](https://tabbyml.github.io/weight-space-playground/)

Previous HN discussion: [https://news.ycombinator.com/item?id=50341174](https://news.ycombinator.com/item?id=50341174)

Comments URL: [https://news.ycombinator.com/item?id=50561436](https://news.ycombinator.com/item?id=50561436)

Points: 1

# Comments: 0
