Skip to content

AI Text to Image

Node icon
AI

Generates images from text prompts using AI, with four model tiers from fast drafts to maximum quality.

AI Text to Image creates images from a text description. Four model tiers balance speed and quality: FLUX Fast (quick iterations), FLUX.2 Pro (professional output), Nano Banana 2 (best prompt understanding, web search), and FLUX.2 Max (highest fidelity).

The output is a standard image that wires into any other node: feed it through Palette Remap, add effects, composite it, or export directly. AI is just another node in the pipeline.

Prompt

PromptText
Range: 4096 chars

Text description of the image to generate

Negative PromptText
Range: 512 chars

Things to exclude (Fast only)

Model

Model TierChoice
Default: Fast

0=Fast, 1=Quality, 2=Advanced, 3=Premium

  • FastFLUX Fast (Schnell) — fastest generation for quick iteration and drafts. Lowest cost.
  • QualityFLUX.2 Pro — high-quality generation for professional output.
  • AdvancedNano Banana 2 — best prompt understanding, optional web search.
  • PremiumFLUX.2 Max — highest fidelity generation from Black Forest Labs.

Generation

Output WidthInteger
Default: 1024 Range: 64–2048

Generated image width (multiple of 8 for Fast, 16 for Quality/Premium)

Output HeightInteger
Default: 1024 Range: 64–2048

Generated image height

StepsInteger
Default: 4 Range: 1–4

Diffusion steps (Fast only, capped at 4)

Guidance ScaleNumber
Default: 3.5 Range: 1.0–20.0

How closely to follow the prompt (Fast only)

Safety ToleranceInteger
Default: 2 Range: 1–5

Content filter strictness (Quality/Premium only)

Aspect RatioChoice
Default: 6 Range: 0–10

Advanced tier aspect ratio

ResolutionChoice
Default: 1 Range: 0–3

Advanced tier: 0.5K/1K/2K/4K

Safety ToleranceInteger
Default: 2 Range: 1–6

Advanced tier content filter strictness

Enable Web SearchToggle
Default: false

Allow web search for context (+1 credit)

SeedInteger
Default: -1 Range: -1–MAX

Random seed; -1=random each time

Notes

Tier-dependent visibility: Fast shows Output Width/Height (with size presets), Steps, Guidance Scale, Negative Prompt. Quality/Premium show Output Width/Height (with size presets, 16px-aligned), Safety Tolerance. Advanced shows Nb2 * params. Size presets available for all per-MP tiers.

  • Start with Fast tier for prompt iteration, then switch to Quality/Advanced for final output.
  • Use a negative prompt to exclude unwanted elements.
  • Wire the output into Remove BG to get a clean cutout on transparent background.
  • Set a specific seed to reproduce a result you like, then tweak the prompt while keeping the same composition.