[Windows ROCm multi-GPU][DynamicVRAM] retained ModelVBAR reservation after full unload causes ~38 GB private-memory plateau
![[Windows ROCm multi-GPU][DynamicVRAM] retained ModelVBAR reservation after full unload causes ~38 GB private-memory plateau](https://www.chat-gpts.plus/wp-content/uploads/2026/10/15993-f0029d9c-768x403.jpg)
在 Windows + ROCm 多卡环境启用 DynamicVRAM 并走到模型“完全卸载”路径后,若出现 [Windows ROCm multi-GPU][DynamicVRAM] retained ModelVBAR reservation after full unload causes ~

![[Bug]: Responses bridge returns a narrated tool call as TWO chat choices (text in choices[0] with finish_reason stop, function_call in choic](https://www.chat-gpts.plus/wp-content/uploads/2026/10/43316-2d1db511-768x403.jpg)
![[Security]: litellm PyPI package (v1.82.7 + v1.82.8) compromised — full timeline and status](https://www.chat-gpts.plus/wp-content/uploads/2026/10/24518-7eaf0753-768x403.jpg)
![[Bug]: `litellm_settings.max_budget` ignores `budget_duration`; global proxy budget is a hardcoded trailing-30-day cap](https://www.chat-gpts.plus/wp-content/uploads/2026/10/31292-3ad334c3-768x403.jpg)



![[RFC] Add `modeling_xxx_fusion.py` to support kernel fusion](https://www.chat-gpts.plus/wp-content/uploads/2026/10/13845-d26e81d5-768x403.jpg)
