"Out of Memory" error with CUDA.

[iamhumanipromise]

torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 2.00 GiB (GPU 0; 7.79 GiB total capacity; 4.95 GiB already allocated; 1.60 GiB free; 5.15 GiB reserved in total by PyTorch) If reserved memory is >>                       
allocated memory try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF 

RTX 2070 8GB MAX-Q

---

Which documentation am I supposed to read to set up appropriate memory management? I can't seem to find it for some reason. Unsure about which settings to change and where.

[WASasquatch]

Easy Diffusion doesn't have the best memory management because it uses uses Diffusers API and they are very wonky about their philosophy on what to do add, leaving most this stuff to users to submit custom pipelines, which often break with diffusers changes. A really bad implementation.

Are you running this local? If so, I suggest trying ComfyUI. Node network is a little overwhelming at first, but it's super fast, and very memory efficient. I can do bigger images in Comfy then I can in WebUI