Model
Trends
.ai
Model
Trends
.ai
les meilleurs modèles d'IA open source

Nœud Conditioning (Combine) and (Concat) dans ComfyUI

Conditioning (Combine) merges two conditionings so the sampler is guided by both at once; Conditioning (Concat) joins the token sequences into one longer prompt. Related nodes set area, mask and timestep ranges.
Pack de nœuds : ComfyUI core
Catégorie : Conditionnement et prompts
Cette page n'est pas encore traduite ; elle est affichée en anglais.

Ce que fait Conditioning (Combine) and (Concat)

ComfyUI treats prompts as data, so they can be mixed after encoding. Conditioning (Combine) takes two CONDITIONING inputs and outputs a list containing both; the sampler evaluates each and averages the guidance, which blends two ideas (for example two styles) more evenly than writing them in one prompt. Conditioning (Concat) appends the second prompt tokens to the first, which behaves like a long single prompt.
The same family includes Conditioning (Average) with a weight slider, Conditioning (Set Area) and (Set Mask) to apply a prompt only to a region, and ConditioningSetTimestepRange to apply a prompt only during part of the sampling. Together they make regional prompting and prompt scheduling possible without custom nodes.

Entrées

Nom
Type
Rôle
conditioning_1
CONDITIONING
First encoded prompt.
conditioning_2
CONDITIONING
Second encoded prompt (Concat: appended; Combine: evaluated alongside).

Sorties

Nom
Type
Rôle
CONDITIONING
CONDITIONING
The merged conditioning for the sampler positive or negative input.
AD
Oubliez l'installation : des workflows ComfyUI préinstallés sur le cloud
BitVector exécute des workflows ComfyUI prêts à l'emploi sur ses propres GPU. Rien à installer, aucun nœud manquant, aucune case rouge. Ouvrez-le sur votre téléphone ou votre portable et générez en une minute.

Comment utiliser Conditioning (Combine) and (Concat)

  1. Encode two prompts with two CLIP Text Encode nodes.
  2. Feed both into Conditioning (Combine) or (Concat).
  3. Connect the output to the KSampler positive input.
  4. For a region, pass one prompt through Conditioning (Set Mask) with a MASK first, then combine with the full-image prompt.

Réglages et conseils

  • Combine doubles the sampler work per step for the positive side; Concat does not.
  • Set Area with strength 1.0 and set_cond_area "default" is the classic two-subjects-left-and-right technique.
  • Use ConditioningSetTimestepRange 0.0-0.3 for a composition prompt and 0.3-1.0 for a detail prompt.
  • Conditioning (Average) with conditioning_to_strength 0.5 gives a true midpoint between two prompts.

Dépannage de Conditioning (Combine) and (Concat)

Subjects bleed into each other with Set Area or Set Mask

Pourquoi cela arrive
The regional prompts overlap or there is no general prompt covering the whole canvas, so the model fills gaps from either region.

Comment le corriger
Add a third, general prompt for the whole image, combine all three, and keep the regions from overlapping. Lower the regional strength to 0.8.

Error: cond shapes do not match when using Concat

Pourquoi cela arrive
The two prompts come from different text encoders (for example SD 1.5 CLIP and SDXL CLIP) or one includes pooled output the other lacks.

Comment le corriger
Encode both prompts with the same CLIP object from the same loader.

Nothing changes when combining

Pourquoi cela arrive
Both inputs are the same conditioning, or the combined output is not the one wired into the sampler.

Comment le corriger
Check the wire into KSampler positive comes from the combine node and that the two inputs differ.

AD
Lancez ce workflow depuis votre téléphone
Chaque workflow de cette page est préinstallé sur BitVector, modèles et nœuds personnalisés déjà en place. Choisissez-en un, tapez un prompt, c'est fait. Zéro configuration, rien à télécharger.

Questions sur Conditioning (Combine) and (Concat)

Combine or Concat for two styles?

Combine blends two separately evaluated guidances; Concat is like a longer prompt. Try Combine first for styles, Concat for added detail words.

Can I do this on the negative side?

Yes, all conditioning nodes work for negative prompts too.

Nœuds liés

AD
Marre de réparer des nœuds ? Laissez faire le cloud
BitVector garde des centaines de workflows ComfyUI installés, à jour et testés sur des GPU cloud rapides. Pas de Python, pas d'erreurs CUDA, pas de limite de VRAM. Depuis n'importe quel navigateur.

Autres nœuds ComfyUI

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.