[Bug]: LoRA `lora_expand` Triton kernel outputs NaN on Hopper (sm_90) with `block_n=128`, producing garbled LoRA output
![[Bug]: LoRA `lora_expand` Triton kernel outputs NaN on Hopper (sm_90) with `block_n=128`, producing garbled LoRA output](https://www.chat-gpts.plus/wp-content/uploads/2026/10/48590-b9235c5a-768x403.jpg)
在 Hopper(sm_90,如 H20/H100)上,用 vLLM 加载 LoRA 适配器时,如果输出全是 !!!!!! 之类的乱码,或者 /v1/completions 返回 Out of range float values are not JSON compliant: nan ,通常是 L




![[Bug]: Failed to add MinerU](https://www.chat-gpts.plus/wp-content/uploads/2026/10/20881-960016ec-768x403.jpg)


![[Bug]: Scale-out token path breaks stop strings: stop text leaks into derendered output (include_stop_str_in_output ignored) and --tokens-on](https://www.chat-gpts.plus/wp-content/uploads/2026/10/57052-93eb6054-768x403.jpg)
![issue: [regression] [dev] someone removed from the group keeps a chat shared with "Allow replies" open again](https://www.chat-gpts.plus/wp-content/uploads/2026/10/32527-73cee0cc-768x403.jpg)