Qwen3.8-Flash-Next's open 180B weights hailed as a model organism for backdoor research
jd_pressman · x · 2026-10-27
Qwen released Qwen3.8-Flash-Next on Hugging Face (Image-Text-to-Text, 180B params, BF16), already at 3.16M monthly downloads. Researcher jdpressman notes 180B is large but loadable for independent researchers — interpretability researchers have been starved for model organisms of backdoors and ambiguously malicious behavior, and these open weights fill that gap. He also cites MIRI's List of Lethalities point 25, asking whether such behavior can even be explained, or whether malicious training could make explanations unreliable.
Related event: Qwen Releases 180B Open Weights, Sparking Backdoor Debate(2 posts)→
More from Models
- Poster surprised that v4.1 flash gets more accurate when asked to control its CoT — dfrsrchtwts · 2026-10-27
- Dev claims Anthropic is baking 'sabotage' into model behavior for 'wrong' engineering — cephaloform · 2026-10-27
- Gemini fails at listing unread emails, tells user to search Gmail manually — gethackteam · 2026-10-27
- Meta's Muse Spark Open goes live on Applied Compute for async RL post-training — ypatil125 · 2026-10-27
- Moonshot curbs hacking ability in Kimi K3.1 Pro, its strongest 3T-class open-weight model — pstAsiatech · 2026-10-27
- Would you want an argumentative LLM that never backs down, even under prompt injection? — GodComplecs · 2026-10-27