Allocation on device

[ALBIHANY]

I'm having this error which i assume indicates that i don't have enough vram, however i'm able to run the FP8 version of the flux- dev and this exact same model on forge webui with no issues at all, so i'm not sure what's going on here i thought comfyUI supposed to be better optimized to than WebUI yet the model run on forge and not on comfy !!!

[ChibiChubu]

Did you put --high/med/low on the note on comfy?

[ALBIHANY]

Did you put --high/med/low on the note on comfy?

If you mean comfyui arguments, yes i use --lowvram argument, but i don't think that has any affect on for that issue and again i can run the F8 version with the same setting it's just slow and i was hoping for the NF version to be faster, also i'm currently running the same NF model file with forge webui and it works smoothly, i think something is mismanaging the memory in comfyui with that model

[ChibiChubu]

Did you put --high/med/low on the note on comfy?

If you mean comfyui arguments, yes i use --lowvram argument, but i don't think that has any affect on for that issue and again i can run the F8 version with the same setting it's just slow and i was hoping for the NF version to be faster, also i'm currently running the same NF model file with forge webui and it works smoothly, i think something is mismanaging the memory in comfyui with that model

it because i have the same issue (i'm using 3090), it only happen when i use LoRA Boreal. I need remove those arguments. Then it work perfectly.

[ALBIHANY]

Did you put --high/med/low on the note on comfy?

If you mean comfyui arguments, yes i use --lowvram argument, but i don't think that has any affect on for that issue and again i can run the F8 version with the same setting it's just slow and i was hoping for the NF version to be faster, also i'm currently running the same NF model file with forge webui and it works smoothly, i think something is mismanaging the memory in comfyui with that model

it because i have the same issue (i'm using 3090), it only happen when i use LoRA Boreal. I need remove those arguments. Then it work perfectly.

i will give it a shot t today and report back if that actually work

[ALBIHANY]

Did you put --high/med/low on the note on comfy?

If you mean comfyui arguments, yes i use --lowvram argument, but i don't think that has any affect on for that issue and again i can run the F8 version with the same setting it's just slow and i was hoping for the NF version to be faster, also i'm currently running the same NF model file with forge webui and it works smoothly, i think something is mismanaging the memory in comfyui with that model

it because i have the same issue (i'm using 3090), it only happen when i use LoRA Boreal. I need remove those arguments. Then it work perfectly.

i will give it a shot t today and report back if that actually work

just tried it, it doesn't work still the same issue