DeAlignAI publishes altered GLM-5.3-Flash weights, claims 320/320 harmful-prompt compliance
Jinho Jang's project made the modified model downloadable and reported compliance with every HarmBench prompt in its own test. The result has no established independent validation.
By RuntimeWire Staff · Published
Primary source: jyn.dev
Why it matters
Open weights let developers inspect and deploy capable models locally while giving downstream users the same ability to modify safeguards. DeAlignAI's downloadable artifact demonstrates that redistribution path, although its self-reported benchmark does not establish real-world offensive capability.
Independent AI researcher Jinho Jang has published altered weights for GLM-5.3-Flash, an open-weight model from Beijing-based Z.ai, that his DeAlignAI project describes as having its guardrails removed. The downloadable artifact shows what a permissive model license allows, even if DeAlignAI's performance claims remain unverified by outside researchers.…
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.