jasperai

/

Flux.1-dev-Controlnet-Upscaler

like 423

Follow

Jasper AI 136

It works with 12 but the main flux model has to be quantized.

@Karachay Thanks for the reply. So I can use quantized Flux?
For example, I can use this model:
https://huggingface.co/lllyasviel/flux1-dev-bnb-nf4
Or does the upscaler model also need to be quantized? There is no such model now?

Can you please explain how quantizing a model works? I have 16 GBs of RAM and I am not able to load the model with the following code:

import torch
from diffusers import FluxControlNetModel
from diffusers.pipelines import FluxControlNetPipeline

# Load pipeline
controlnet = FluxControlNetModel.from_pretrained(
  "jasperai/Flux.1-dev-Controlnet-Upscaler",
  torch_dtype=torch.bfloat16
)
pipe = FluxControlNetPipeline.from_pretrained(
  "black-forest-labs/FLUX.1-dev",
  controlnet=controlnet,
  torch_dtype=torch.bfloat16
)
pipe.to("cuda")

I get this error:
OutOfMemoryError: CUDA out of memory.

@Karachay

PROBLEM SOLUTION:

import torch
from diffusers.utils import load_image
from diffusers import FluxControlNetModel, BitsAndBytesConfig, FluxTransformer2DModel
from diffusers.pipelines import FluxControlNetPipeline


nf4_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.float16
)

controlnet = FluxControlNetModel.from_pretrained(
    "jasperai/Flux.1-dev-Controlnet-Upscaler",
    quantization_config=nf4_config,
)

model_id = "black-forest-labs/FLUX.1-dev"
nf4_id = "sayakpaul/flux.1-dev-nf4-with-bnb-integration"
model_nf4 = FluxTransformer2DModel.from_pretrained(nf4_id, torch_dtype=torch.float16)

pipe = FluxControlNetPipeline.from_pretrained(
    model_id,
    transformer=model_nf4,
    torch_dtype=torch.float16,
    controlnet=controlnet
)
pipe.enable_model_cpu_offload()

control_image = load_image(
    "image.jpg"
)

image = pipe(
    prompt="", 
    control_image=control_image,
    controlnet_conditioning_scale=0.6,
    num_inference_steps=28, 
    guidance_scale=3.5,
    height=control_image.size[1],
    width=control_image.size[0]
).images[0]
image.save("upscaled_img_quanted.png")

@indianspice @andchir if you have not solved it yet