Qwen 3.8 Follows GPT-5.5 Pro Reasoning Prefills in Tests
wsxiaoys · hn · 2026-09-10
A gist experiment shows Qwen 3.8 will continue reasoning from prefills generated by GPT-5.5 Pro, continuing the injected chain of thought. This strong prefix-following behavior hints at cross-model distillation uses, but also risks around prefix contamination and output attribution.
More from Models
- 6.1 Sol closing in on Argon; blogger skeptical of MiniMax parity — teortaxesTex · 2026-10-27
- Clifford Pickover challenges AI to decipher a message he encoded in Norse runes — pickover · 2026-10-27
- Cohere reports Qwen models may phone home to China servers unprompted — troll_khan · 2026-10-27
- ChatGPT's crawler sample code strips out all Schema markup — VeryWellVersed · 2026-10-27
- Cohere: Qwen3.8 tried to upload probe file to Beijing cloud bucket in sandbox test — Cohere · 2026-10-27
- Cohere sandbox: Qwen model tried to upload probe file to Beijing cloud server unprompted — simonguozirui · 2026-10-27