Flux Guidance node in ComfyUI
Flux Guidance (FluxGuidance) writes the guidance value into the conditioning that Flux Dev was distilled to expect, replacing cfg; 3.5 is the default and 2 to 5 the useful range.
Node pack: ComfyUI core
Category: Conditioning and prompts
What Flux Guidance does
Flux Dev is a guidance-distilled model: instead of running the prompt twice per step like classic CFG, it reads a single guidance number from the conditioning. Flux Guidance sets that number. The sampler cfg should stay at 1.0, which means no classic CFG and no negative prompt cost.
Lower guidance (1.5-2.5) gives softer, more photographic and more varied images; higher values (4-6) follow the prompt harder with more contrast and the typical plasticky Flux look. Flux Schnell ignores the value, and Flux Kontext and Flux Fill also use it.
Inputs
Name | Type | What it is |
|---|---|---|
conditioning | CONDITIONING | The encoded positive prompt from CLIP Text Encode. |
guidance | FLOAT | The distilled guidance strength. 3.5 default, 2-5 typical. |
Outputs
Name | Type | What it is |
|---|---|---|
CONDITIONING | CONDITIONING | The prompt with guidance embedded, for the sampler positive input. |
AD
Skip the setup: ComfyUI workflows preinstalled on the cloud
BitVector runs ready-made ComfyUI workflows on its own GPUs. No install, no missing nodes, no red boxes. Open it on your phone or laptop and generate in a minute.
How to use Flux Guidance
- Encode the prompt with CLIP Text Encode (CLIP from DualCLIPLoader).
- Pass the CONDITIONING through Flux Guidance.
- Connect the result to KSampler positive; set KSampler cfg to 1.0.
- For the negative input, use ConditioningZeroOut of the same prompt or an empty prompt.
Settings and tips
- For realism try guidance 2.0-2.5 with 25-30 steps; for illustration and text rendering 3.5-4.5.
- Guidance above 6 gives burnt skin and oversharpening.
- If you want a real negative prompt on Flux, raise cfg to 1.5-2 and expect roughly double the time per step.
- Flux Kontext editing usually works best near 2.5-3.0.
Troubleshooting Flux Guidance
Flux images are washed out and blurry
Why it happens
No Flux Guidance node in the chain, so the model runs at an effective guidance of 1, or cfg is also 1 with no guidance embedded.
How to fix it
Add Flux Guidance between the prompt and the sampler, set it to 3.5.
Images look plastic, oversaturated or have grid artifacts
Why it happens
Guidance is too high, or classic cfg above 1 is combined with distilled guidance.
How to fix it
Keep cfg at 1.0 and lower guidance to 2.5-3.5.
The value does nothing on Flux Schnell
Why it happens
Schnell is a timestep-distilled model that does not read the guidance value.
How to fix it
Expected. Use Flux Dev when you need guidance control.
AD
Run this workflow from your phone
Every workflow on this page is preinstalled on BitVector with the models and custom nodes already in place. Pick one, type a prompt, done. Zero setup, nothing to download.
Questions about Flux Guidance
Should cfg be 1.0 with Flux?
Yes, for Dev and Schnell. The guidance number replaces it.
Does Flux Guidance apply to the negative prompt?
No, only to the positive conditioning. The negative is normally ignored at cfg 1.
Related nodes
Load Diffusion Model
ComfyUI core
Load Diffusion Model (UNETLoader) loads a bare denoising network such as Flux, SD3.5, Wan or HunyuanVideo from models/diffusion_models, with a weight_dtype option for fp8 to save VRAM.
DualCLIPLoader
ComfyUI core
DualCLIPLoader loads two text encoders at once, for example clip_l plus t5xxl for Flux or clip_g plus t5xxl for SD3, and outputs one CLIP object for the prompt nodes.
CLIP Text Encode (Prompt)
ComfyUI core
CLIP Text Encode turns a text prompt into CONDITIONING using the model text encoder. One node holds the positive prompt, a second one the negative prompt.
KSampler
ComfyUI core
KSampler runs the denoising loop: it takes the model, prompts and a latent and produces the finished latent image, controlled by seed, steps, cfg, sampler, scheduler and denoise.
EmptySD3LatentImage
ComfyUI core
EmptySD3LatentImage creates the 16-channel empty latent that Flux, SD3 and SD3.5 need; it replaces Empty Latent Image in those workflows.
AD
Tired of fixing nodes? Let the cloud do it
BitVector keeps hundreds of ComfyUI workflows installed, updated and tested on fast cloud GPUs. No Python, no CUDA errors, no VRAM limits. Works from any browser.
More ComfyUI nodes
ComfyUI Manager
ComfyUI-Manager
ComfyUI Manager is the extension that installs, updates and fixes custom node packs and models from inside the interface, resolves missing nodes in imported workflows and snapshots your setup.
Load Checkpoint
ComfyUI core
Load Checkpoint (CheckpointLoaderSimple) opens a .safetensors or .ckpt model file and hands out the three parts every workflow needs: the diffusion MODEL, the CLIP text encoder and the VAE.
Save Image
ComfyUI core
Save Image writes the IMAGE tensor to ComfyUI/output as a PNG, with the whole workflow embedded in the file metadata so the picture can be dragged back into ComfyUI to restore the graph.
VAE Decode
ComfyUI core
VAE Decode converts the sampled LATENT into a pixel IMAGE with the VAE; VAE Decode (Tiled) does the same in tiles for very large images.
Empty Latent Image
ComfyUI core
Empty Latent Image creates the blank latent canvas (width, height, batch_size) that text-to-image sampling starts from; dimensions must be multiples of 8 and match the model family.
Load Image
ComfyUI core
Load Image reads a picture from the ComfyUI input folder (or an upload) and outputs an IMAGE tensor plus a MASK taken from the alpha channel, the starting point for img2img, inpainting, ControlNet and IP-Adapter graphs.
Model
Trends
.ai
© ModelTrends.ai
|
Made in Japan
|
© 2026