As of: 2026-11-09 21:13 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/p/2026-11-02-rikiyatakehi-astra-edits-qwen-weights/

# GPT-6 Astra edited Qwen3.5-0.8B weights (0.016%) to fix reasoning loops

Rikiya Takehi (@rikiyatakehi), X, 2026-11-02. Importance: major (4 of 5).
Source: https://x.com/rikiyatakehi/status/2117313742891327994

## Why it matters

First-hand announcement of weight-space programming; ~6.5M views (fxtwitter, Nov 3).

## Summary

Lead author announces that GPT-6 Astra directly changed 0.016% of Qwen3.5-0.8B's weights to fix its reasoning loops and improve maths. About 6.53M views when read via fxtwitter on Nov 3.

## Archived text

> 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: https://rikiyat.github.io/blog/weight-space-programming/

Archived 2026-11-04 via fxtwitter (unofficial).
Counts: 9,219,641 views, 55,522 likes, 7,268 reposts, 2,323 replies (at fetch time)
Media: https://pbs.twimg.com/media/HWI1oKUWUAAWoZp.jpg?name=orig

## Cited in

- 2026-11-02: [MIT CSAIL researchers](https://postcutoff.com/e/2026-11-02-gpt-6-astra-edits-qwen-weights-weight-space-programming/)