Leonardo's highly adaptable, prompt-responsive image model
Lucid Origin from Leonardo is the model that follows instructions most literally. There is nothing to configure beyond the prompt, so the way to steer it is to write more precisely — which makes it a good place to learn what your prompt is actually asking for.
Written for this model in particular. Open one in the composer and edit it from there.
“A cluttered watchmaker's bench seen from above: loupe, tweezers, three open movements, brass filings, warm lamp from the upper left, shallow depth of field”
Try this prompt“A storm-lit prairie with a single red barn at centre-left, wheat bending right, dark anvil cloud above”
Try this prompt“A cross-section illustration of a submarine, labelled compartments, technical drawing style”
Try this promptOpenAI's image model, and the one to trust with text in the frame
GoogleGoogle's highest-fidelity Gemini image model, up to 4K
GoogleGoogle's fastest Gemini image model
RecraftFast, cost-efficient image generation with real style controls
RecraftHigh-resolution Recraft, at 2048px and above
RecraftThe only model here that outputs production-ready SVG
50 free credits on sign-up, no credit card required.