GPT Image·par OpenAIText to image
GPT Image 2.5 Flare Text-to-Image
Image generation from natural-language prompts with OpenAI GPT Image 2.5. 14 sizes up to 4K (3840×2160), five quality tiers including xhigh and max, and true transparent backgrounds. Prompts up to 32,000 characters. Priced live from quality, size and prompt length.
openai/gpt-image-2.5-flare/text-to-imageParamètres
| Paramètre | Type | Par défaut | Plage ou options |
|---|---|---|---|
Output format output_formatOutput image format. PNG keeps transparency; JPEG files are smaller. | select | png | png, jpeg |
Background backgroundtransparent produces a real alpha channel (needs PNG output). Set it here — asking for a transparent background in the prompt paints a checkerboard. | select | auto | auto, opaque, transparent |
Moderation moderationContent-moderation strictness on the provider side. low applies less restrictive filtering. | select | low | low, auto |
Quality qualityRendering quality tier. xhigh and max are new in GPT Image 2.5. At 1024×1024: low ≈0.6×, high ≈3.2×, xhigh ≈5.4×, max ≈11.9× the medium price. | select | medium | low, medium, high, xhigh, max |
Size sizeOutput size in pixels — the only control over dimensions and aspect ratio. Sizes above 2560×1440 are experimental upstream. Larger sizes cost more. | select | 1024x1024 | 1024×1024 · 1:1 (1024x1024), 1024×768 · 4:3 (1024x768), 768×1024 · 3:4 (768x1024), 1536×1024 · 3:2 (1536x1024), 1024×1536 · 2:3 (1024x1536), 2048×2048 · 1:1 (2048x2048), 2048×1152 · 16:9 (2048x1152), 1152×2048 · 9:16 (1152x2048), 2560×1088 · 21:9 (2560x1088), 1088×2560 · 9:21 (1088x2560), 2880×2160 · 4:3 (2880x2160), 2160×2880 · 3:4 (2160x2880), 3840×2160 · 16:9 (4K) (3840x2160), 2160×3840 · 9:16 (4K) (2160x3840) |
Entrées
Aucun fichier de référence : ce modèle travaille à partir du prompt seul.
Guide de prompt pour GPT Image
TARGET MODEL: GPT Image 2 (OpenAI). Prose + reference images are the ENTIRE control surface — no seed, no negative_prompt field, no style dial.
Règles à respecter
- Prose, not tags. Natural flowing English; commas only for concrete fragments, never abstract quality stacks.
- Ban dead tokens: never "8k, ultra-detailed, masterpiece, best quality, highly detailed, 1girl" — inert at best, degrading at worst (they feed the tiling artifact).
- Order: scene/background → subject → key details → constraints. Constraints ALWAYS last. Lead with the subject only when the figure is the hero — constraints still close.
- Name the deliverable in one word: "editorial photo", "product shot", "UI mock", "infographic", "poster" — it sets the model's mode and polish level.
- Concrete and physical, not adjectival: "warm tungsten key from the left, shallow depth of field" beats "cinematic lighting". A lens is a feel cue only ("50mm feel") — the model runs no optical simulation.
- Affirmative beats negative. Describe what IS, materially and confidently. Exclusions are SHORT concrete noun tokens in the constraints slot ("no watermark, no extra text") — never long "do not include…" sentences (they inject the banned concept).
- Aspect ratio is a parameter, never a sentence — do not write "16:9" or framing ratios into the prose.
- Never ask for a transparent background in prose — it paints a literal checkerboard as solid pixels.
- Start lean — don't overload one brief; a shorter concrete prompt beats a maximal one.
Ce qui fonctionne le mieux
- Candid/authentic register — describe the DEGRADATION, not the production: "photorealistic" + one of "candid photograph / amateur photograph / iPhone photo", then the stack: "amateur composition, no studio lighting, subtle film grain, slight sensor noise in darker areas, slight overexposed highlights, mild compression artifacts, imperfect framing, natural color balance". Close on subtraction: "No glamorization, no heavy retouching. No cinematic grading, no flash, no studio light." Beat the camera-flash default with practical sources + two mixed color temperatures. Kill the influencer face: "no beauty filter, no HD polish".
- Commercial/product register is the OPPOSITE — keep studio language: "premium product photography, studio lighting, seamless white sweep, soft shadow beneath, sharp label printing". Pick the register from the stated use.
- Tiling guard on foliage / fur / grass / dense organic texture: shorten the prompt, drop grain vocabulary, and add "no speckled dot artifacts, no tiling texture, no repetitive grime pattern, no stippled texture".
- Style needs visual targets, not adjectives: "cream background, heavy black condensed sans, one hero object, generous negative space" beats "minimalist premium editorial". Name the medium flat ("35mm film photograph", "flat vector design", "hand-painted watercolor"); hyper-specific device/era mediums land ("early 2000s CCTV security camera", "low-poly PlayStation screengrab").
- Fight the warm/yellow house cast with explicit "natural color balance".
- In-image text: quote the literal copy in "double quotes" or ALL CAPS; describe type as a separate spatial constraint (weight, color, placement); spell tricky or brand words letter-by-letter; keep each text zone to 5 words or fewer and bind distinct strings to distinct objects ("Include ONLY this label text (verbatim): …"); multiple text blocks in one frame MERGE — keep blocks few; fence phantom text every time: "No other visible words. No random letters. No duplicate text. No watermark." and assert the single instance ("Render the tagline exactly once.").
- Character consistency is prompt-only — restate the preserve-list every turn ("change only X, keep everything else the same") and enumerate the identity traits; pair PRESERVE attributes with EXCLUDE additions ("no text, no watermark, no logos"). One change per edit.
Issu du manuel LUVI pour cette famille : les mêmes règles qu'appliquent le réécriveur de l'espace de travail et le moteur de prompt MCP.