flux-2-klein-4b_clear

tumbnail_image_anime
tumbnail_image_photoreal

A fine-tuned version of flux-2-klein-4b model, designed to produce clearer and more vibrant outputs.

Compare to flux-2-klein-4b

flux-2-klein-4b_clear_compare_anime
flux-2-klein-4b_compare_anime

flux-2-klein-4b_clear | flux-2-klein-4b (Original)

flux-2-klein-4b_clear_compare_photoreal
flux-2-klein-4b_compare_photoreal
compare_photoreal_close_up

Compared to the original, flux-2-klein-4b_clear generates illustrations with higher contrast and richer colors.


INT8_ConvRot_HQ (light weight)

INT8_ConvRot_HQ is a lightweight version.

Image quality is slightly lower than the original BF16, but it is particularly well-suited for fast generation in low-VRAM environments (4-6 GB).


Workflow

Text to Image

text_to_image_workflow

This PNG image contains a ComfyUI workflow.

Image Edit

image_to_image_workflow

Recommended Settings

General

  • Sampler:
    • euler (Speed)
    • heunpp2 (Quality)
  • Steps: 12 (4-15)

Upscaler

your choice, examples:


Tips (External site)


License

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support