Saltar al contenido
WorkflowPublicadoExterno
Abrir fuente

-DepriveD- Flux 2 Klein 9B T2I Workflow [Built-in Creator]

DepriveD FLux 2 Klein 9B – Simple All-in-One Text-to-Image Workflow

VersiónRelease 3226671 — DeprivedFlux2Klein9BT2I_v10.jsonVRAM estimadaNot yet characterizedDcalidad

Perfil de ejecución

Acceso
Flujo externo
Estado
Publicado
Versión
DeprivedFlux2Klein9BT2I_v10.json
Fuente
Civitai
Actualizado
16 oct 2026

Descripción de la fuente

DepriveD FLux 2 Klein 9B – Simple All-in-One Text-to-Image Workflow

A clean, beginner-friendly, one-stop workflow for Flux 2 Klein 9B. Press a few buttons → hit Queue (or just Ctrl + Enter ) → get great images.

What this workflow is

This is designed to be simple and fast. Everything important is clearly labeled and color-coded so you can understand it quickly. No messy spaghetti.

Built-in Prompt Creators (Free)

I made three custom prompt creators that are included:

Scene Creator → Generates a random scene with random assets

Creature Creator → Generates a random colossal creature in a random setting

Character Creator → Generates a random character in a random setting

Only use ONE Creator a a time. When a creator is turned on, you don’t need a positive prompt . Together they give tens of thousands of possible combinations (before even changing the seed). You can edit them later if you know how.

Optional AI Prompt (LLM)

There is an optional LLM that can flesh out your prompts. You can use it:

With a normal positive prompt

Together with the Creators

Or by itself

It is not required .

Required Custom Nodes

ComfyUI-Pixaroma

rgthree-comfy (for reroute nodes)

Install them through ComfyUI Manager if you don’t already have them.

Important Notes

Images do NOT save automatically. On the Compare Image node, click Show and then Save manually.

Sage Attention is enabled for speed. If you have issues, turn it off.

Recommended Steps and CFG are shown in the General Info panel on the left.

How to use (super simple)

Choose your resolution

Set Steps & CFG (or leave the recommended settings)

(Optional) Turn on AI Prompt

Choose an Art Style if you want

Write a prompt or turn on one of the Creators

(Optional) Add LoRAs

Hit Queue or just press Ctrl + Enter

Required Models

Flux-Klein-2-9B-INT8 ConvRot

Qwen_3_8b_fp8mixed (CLIP)

Full_Encoder_Small_Decoder VAE

Dolphin-2.9-Llama3-8B-Q4_K_M (for the optional AI Prompt)

A system prompt for the LLM is also included in the download.

About the LoRAs

The LoRAs currently loaded in the panel were not used for any of the example images. They are just placeholders. Feel free to replace them with whatever you want.

HAPPY CREATING!!!!

What you need

Model files
0.1 GiB known subtotal

Remaining hardware or download-size estimates are not available yet.

Requisitos

Modelos

7
  • 4x_foolhardy_Remacri (1).safetensors

    Posible

    Upscaler · Unknown

  • dolphin-2.9-llama3-8b-Q4_K_M.gguf

    No resuelto

    Checkpoint · Unknown

  • flux-2-klein-9b-int8-ConvRot-comfyui.safetensors

    Posible

    Checkpoint · Unknown

  • Checkpoint · 12.6 GB · SAFETENSORS · bf16 · full · Unknown

  • qwen_image_vae.safetensors

    Archivo sin verificar

    VAE · Hugging Face · Comfy-Org/Krea-2

  • Text encoder · 4.88 GB · SAFETENSORS · Hugging Face · Comfy-Org/Krea-2

  • Upscaler · 64.0 MB · PTH · Unknown

Paquetes de nodos

8

Fuentes

1 fuente
CivitaiPrincipalVersión 3226671
civitai.com/models/2856825

Verificado 16 oct

Extractos de fuentes

1 extracto

Source context: 175 downloads · Type Workflows · Base model Flux.2 Klein 9B