My stuff

  • My Workflows

  • Liked Workflows

  • Following Workflows

Go to OpenArt main site
Upload workflow

Nunchaku Flux Kontext vs. Turbo LoRA

5.0

0 reviews
6
3.7K
1.3K
0
Description

Unlock the full potential of the Flux Kontex Dev model without the agonizing wait! This workflow solves the biggest pain point of this powerhouse model: slow generation times. By leveraging Nunchaku technology, you get 8x faster image generation (as low as 5 seconds on an RTX 4090) while preserving near-identical quality to the original model. Perfect for rapid prototyping, testing prompts, or batch generation!

YouTube Tutorial:  

https://youtu.be/pTSKfLB8b_E

Discussion

(No comments yet)

Loading...

Author

23
236.7K
1.6K
599.4K

No reviews yet

  • - latest (2 months ago)

Primitive Nodes (11)

Anything Everywhere3 (2)

Fast Groups Muter (rgthree) (1)

FluxGuidance (1)

FluxKontextImageScale (1)

Note (1)

NunchakuFluxDiTLoader (1)

Prompts Everywhere (1)

ReferenceLatent (1)

UnetLoaderGGUF (2)

Custom Nodes (17)

ComfyUI

  • - VAEEncode (1)

  • - ConditioningZeroOut (1)

  • - VAELoader (1)

  • - DualCLIPLoader (1)

  • - KSampler (3)

  • - CLIPTextEncode (1)

  • - LoadImage (1)

  • - LoraLoaderModelOnly (1)

  • - VAEDecode (3)

  • - PreviewImage (3)

ComfyUI Easy Use

  • - easy seed (1)

Checkpoints (0)

LoRAs (1)

flux/FLUX.1-Turbo-Alpha.safetensors