OrangeMix & MixPro Notebooks Disconnects Under 10 Minutes Consistently

[nuroisea]

OrangeMix & MixPro Notebooks Disconnects Under 10 Minutes Consistently

[DigKingy]

Note: The LoRa's I have loaded were not even registered within the Notebook, leaving no effect towards the diffusion except almost immediately disconnect the notebook due to me trying to put the LoRa's in effect.

[NUROISEA]

Error while deserializing header: MetadataIncompleteBuffer

What LoRAs are you using? I assume that you are downloading it via Batchlinks? I tested it with my favorite LoRAs, and all of which works, probably a problem on your downloaded LoRAs.

---

About the 10 minutes, yeah I experience it too for some reason. This is probably Colab's way of limiting usage, I have no control over this.

[DigKingy]

LoRA's: lora:Rabbit (wlsdnjs950):0.8, lora:aki:0.3, lora:3DMM_V11:0.5, lora:add_detail:0.6

These are the LoRA's I would use in order to diffuse different pieces of media. Yet, maybe around 2 minutes of implementing them into the system, it would give me the error below. 👇

"loading Lora /content/stable-diffusion-webui/models/Lora/3DMM_V11.safetensors: SafetensorError
Traceback (most recent call last):
File "/content/stable-diffusion-webui/extensions-builtin/Lora/lora.py", line 253, in load_loras
lora = load_lora(name, lora_on_disk)
File "/content/stable-diffusion-webui/extensions-builtin/Lora/lora.py", line 162, in load_lora
sd = sd_models.read_state_dict(lora_on_disk.filename)
File "/content/stable-diffusion-webui/modules/sd_models.py", line 250, in read_state_dict
pl_sd = safetensors.torch.load_file(checkpoint_file, device=device)
File "/usr/local/lib/python3.10/dist-packages/safetensors/torch.py", line 259, in load_file
with safe_open(filename, framework="pt", device=device) as f:
safetensors_rust.SafetensorError: Error while deserializing header: MetadataIncompleteBuffer"

But that's the thing, it doesn't screw up until I add the LoRA's, but if I just use the model as is, no problems occur.