HeyGen Image 1
Use it ↗Puts 1 to 5 photos of one person into a single-person reference scene: 1 to 5 images per run, up to 3840 x 2160. Best for portraits in a set scene; scene pose and clothing stay.
Takes 1 to 5 photos of the same person and one scene image with exactly one person, then puts that person's face, hair and expression into the scene. Pose, clothing, light and setting come from the scene; there is no text prompt. 1 to 5 images per run, ten aspect ratios from 1:1 to 16:9, seed for repeatable sets. Outputs reached 3840 x 2160 at 16:9 and 2160 x 2880 at 3:4 in our tests. The face in the scene must be large and clearly visible. Best for: placing a specific person into a prepared scene, character consistency across scenes, hero portraits and set portraits. Not for: scenes with several people or a small face (a wide platform shot failed); changing pose or clothing by instruction. Set the aspect ratio explicitly, because empty sometimes returned bars. Instead: edits by written instruction, GPT Image 2.5 Sunburst; up to 14 references, Nano Banana 2.1.