suede299

补充一下。用JavTubeId搜一切正常

`activating extra network lora with arguments [, ]: RuntimeError Traceback (most recent call last): File "D:\suede\Desktop\SDWegUI\modules\extra_networks.py", line 75, in activate extra_network.activate(p, extra_network_args) File "D:\suede\Desktop\SDWegUI\extensions-builtin\Lora\extra_networks_lora.py", line 23, in activate lora.load_loras(names, multipliers)...

The IPAdapter's ATTENTION MASK is the reference portion of the target. For lora to work like this, it needs to be targeted to the generated image, but the images are...

[TensorRT-LLM] TensorRT-LLM version: 0.9.0 9.3.0.post12.dev1  Uses your commits.

> @suede299 see your json/yaml config， is the structure changed？ Is it the .json under RAG\trt-llm-rag-windows-main\config? Maybe I don't know how to modify it  Went back through the TensorRT-LLM...

> @suede299 The NV likes change the config file of the trt engine. When you generate the trt engine files, you will get a file of config.json and rank0.engine. Like...

> @suede299 TensorRT is better for docker server, or edge device, not suitable for consumer clients. The tensorrt model need to keep the same version with TensorRT Library. Every version...

Use [ComfyUI-Inspire-Pack](https://github.com/ltdrdata/ComfyUI-Inspire-Pack)'s global seed node, I guess. If you adapted other UI's to display the current image used seed.

> um, what? I did nothing. I didnt even add any custom nodes. I'm using a pure comfuyi git pull, as managed by stableswarmui. > > Are you saying this...

I had a similar problem, confirmed it was a memory footprint issue, then disabled xformers and replaced it with pytorch 2.1 and the problem went away. In my case with...