{ "model": "openai/gpt-image-2", "model_with_references": "openai/gpt-image-2/edit", "meleon_model": "gpt-image-2", "input": { "num_images": 1, "output_format": "png" }, "stages": { "draft": { "quality": "low", "image_size": { "width": 720, "height": 1280 }, "why": "cheap rough covers to agree the subjects, the empty title band and the set look; 720 x 1280 is the smallest 9:16 size fal accepts (at least 655,360 pixels)" }, "final": { "quality": "medium", "image_size": { "width": 1088, "height": 1936 }, "why": "the art for the finished cover; composed down to 1080 x 1920, so nothing is upscaled" } }, "max_references": 3, "approx_cost": { "per_1024_square": { "low": 0.006, "medium": 0.05, "high": 0.2 }, "per_reference_image": 0.01, "note": "GPT Image 2 is token-billed (image output $30 per 1M tokens, image input $8 per 1M, text $5 in and $10 out per 1M; fal model page read 2026-11-09). The scripts scale fal's per-1024-square figures by pixel count, so a 720 x 1280 draft at low is about $0.005 and a 1088 x 1936 final at medium about $0.10, plus about $0.01 per reference image. fal's published portrait prices (1024 x 1536: low $0.005, medium $0.042) suggest tall images bill a little under this, so the estimate leans high. fal returns no token count with the image; the fal dashboard shows the real charge." }, "empty_space": { "top": "Composition: the band from about 15 to 40 percent of the height, measured from the top, is calm and uncluttered (plain backdrop or soft out-of-focus background, no objects, no busy pattern): a title goes there later. The main subject sits centred in the middle and lower part of the frame, between about 42 and 85 percent of the height.", "middle": "Composition: the band from about 23 to 57 percent of the height, measured from the top, is calm and uncluttered (plain backdrop or soft out-of-focus background, no objects, no busy pattern): a title goes there later. The main subject sits centred low in the frame, between about 60 and 86 percent of the height.", "bottom": "Composition: the band from about 40 to 65 percent of the height, measured from the top, is calm and uncluttered (plain backdrop or soft out-of-focus background, no objects, no busy pattern): a title goes there later. The main subject sits centred high in the frame, between about 15 and 38 percent of the height." }, "grid_rule": "Keep everything important between 13 and 87 percent of the height, because the profile grid shows only that middle part, and keep the right-hand 14 percent of the frame free of important detail.", "presets": { "bold-flat": { "use_for": "food, products, objects, tips with one hero thing per episode", "look": "one hero subject shot from directly above on a saturated flat {{colour}} paper backdrop, hard midday sunlight from the upper left casting one crisp shadow, minimal props, glossy true-to-life styling, high-contrast editorial look, fine natural film grain", "palette": ["tomato red", "butter yellow"], "style": { "treatment": "block", "ink": "#141414", "accent": "#ffffff", "position": "top", "align": "left" } }, "moody-film": { "use_for": "vlogs, stories, travel, behind the scenes", "look": "a cinematic low-key scene, deep shadows and one warm practical light source, muted teal and amber colour grade, 35mm film still with soft halation and visible grain, shallow depth of field", "palette": [], "style": { "treatment": "shadow", "ink": "#f6efe4", "accent": "#e8a54b", "position": "top", "align": "left" } }, "soft-pastel": { "use_for": "lifestyle, wellness, home, how-to series", "look": "a soft pastel {{colour}} seamless studio set, gentle diffused window light from the left, rounded clay-like props in matching tones, clean minimal styling, calm premium lifestyle look", "palette": ["peach", "mint", "lilac"], "style": { "treatment": "plain", "ink": "#2a2233", "accent": "#2a2233", "position": "top", "align": "center" } }, "paper-collage": { "use_for": "explainers, education, opinion, myths and facts", "look": "a handmade cut-paper collage illustration on a {{colour}} textured card background, layered torn paper shapes with soft drop shadows, halftone dots and pencil marks, warm playful editorial illustration style", "palette": ["cream", "sky blue"], "style": { "treatment": "block", "ink": "#ffffff", "accent": "#1d1d1f", "position": "top", "align": "left" } }, "risograph": { "use_for": "music, culture, events, lists, any series that wants a print look", "look": "a two-colour risograph print look in fluorescent pink and deep blue on off-white paper, slight misregistration, grainy ink texture, bold simplified shapes, graphic poster style", "palette": [], "style": { "treatment": "plain", "ink": "#1f2a7a", "accent": "#ff4f9a", "position": "top", "align": "left" } } }, "cost_note": "A set of 6 is about $0.03 for low drafts and about $0.71 for medium finals (each final carries cover 1 and its own draft as references), about $0.74 in all, up to about $0.99 with two re-rolls. build-requests.mjs --stage estimate prints the total for the real plan; quote it and get one yes before the first paid call. The titles, the grid preview and the checks are free and local." }