HongyuChen

After further debugging, we found that this error was caused by the difference between the calculation result and the tensor_ref_d in some of the values. We've added the following code:...

@wenlei-bao Hello, thanks for your advice first. `CUDA_DEVICE_MAX_CONNECTIONS=1` did help me a lot. And I still have a question, which is also about resources competition. I'm curious if FLUX has...