ai-search 5 months ago

New Open-Source Nano Banana Dropped | Light and Ultra-Fast

Explore Flux 2 Klein, a new open-source AI image model, boasting impressive speed and low VRAM requirements.

This tutorial provides a step-by-step installation guide using ComfyUI, demonstrating both image generation and editing capabilities. Learn to leverage various model variants and settings for optimal results.


https://bfl.ai/models/flux-2-klein


Models:

https://huggingface.co/Comfy-Org/flux2-klein-4B/tree/main/split_files

https://huggingface.co/Comfy-Org/flux2-klein-9B/tree/main/split_files

https://huggingface.co/collections/black-forest-labs/flux2


Text to Image Workflow:

https://raw.githubusercontent.com/Comfy-Org/workflow_templates/refs/heads/main/templates/image_flux2_klein_text_to_image.json


Image Editing Workflow:

https://raw.githubusercontent.com/Comfy-Org/workflow_templates/refs/heads/main/templates/image_flux2_klein_image_edit_4b_distilled.json


Guffs:

https://huggingface.co/unsloth/FLUX.2-klein-4B-GGUF

https://huggingface.co/unsloth/FLUX.2-klein-9B-GGUF


00:00 - Flux 2 klein intro

00:34 - Flux2 klein specs and demos

02:48 - Flux2 klein vs Qwen image vs Z image turbo

04:06 - Long text

04:53 - Clock & wineglass

05:18 - Other tricky prompts

05:52 - Other art styles

07:08 - Anatomy

07:34 - UI design

08:00 - Prompt adherence

08:58 - Chinese painting

09:43 - Flux2 Klein vs Qwen Image Edit

10:52 - Changing camera angle

11:59 - Pose transfer

13:23 - How to install Flux2 Klein

14:00 - Mago

15:20 - Installation continued

20:04 - Text to image workflow

21:37 - Image editing workflow

25:49 - Multiple input images

28:23 - For low VRAM text to image

31:58 - Low vram for image editing








AI Search
ai never sleeps!