Generating an image purely from a written description. The model synthesizes composition, subject, lighting and style from the prompt alone, with no input image.
Text-to-image is the base mode of modern image generation: you describe a scene in words and the model renders it from learned visual knowledge — no source photo, no template. Everything about the output (subject, composition, light, style) is steered through the prompt and the generation settings.
Its known weakness is identity: text-to-image alone cannot keep the same face across generations, because every prompt is re-interpreted from scratch. That is fine for one-off scenes and useless for a persona-based account — which is exactly the gap model training closes by moving identity out of the prompt and into the model.
On InfluencerForge.app, text-to-image is what powers photoshoots, with Forge Engine routing each request to the right internal pipeline: your trained persona supplies the identity, the Forge Style supplies the look, and your prompt only has to describe the situation.
Related terms
Generating a new image using an existing image as input alongside a text prompt. The input image anchors composition, pose or subject while the prompt steers what changes.
The text instruction describing what a generation should depict — scene, wardrobe, mood, composition. With a trained persona, prompts describe the situation only; the identity comes from the model.
The practice of writing and systematically refining prompts to get reliable, repeatable results from a generative model — concrete language, deliberate structure, and one change at a time.