4.545 modelos · atualizado em out. de 2026
Seedream 4.5 é o modelo de imagem da ByteDance para pôsteres, fotos de produtos e ilustrações com textos curtos na imagem. No GPT Image Hub, ele renderiza em 2K ou 4K pelos mesmos 33 créditos, com até 10 imagens de referência.
Comece com o guia de escrita abaixo e depois escolha um dos 4.545 modelos da biblioteca, configurados para o Seedream 4.5 e o resto da família Seedream (4.0, 5.0, 5.0 Pro e 5.0 Lite). Cada modelo abre no gerador com a versão recomendada do Seedream já selecionada.
Collectible art toy character design sheet, front, side, and back views. Fluffy, cute character with a round face, wearing [a yellow safety helmet] and [blue round glasses]. Soft tactile plushie figure texture, cute 3D vinyl toy style. Pastel color scheme of [black, cream, and light yellow]. Clean solid beige background, soft studio lighting.
Tang-style fantasy water-element xianxia character poster, [9:16 vertical] aspect ratio. A [23-year-old] oriental female, Tide Sound Pavilion Envoy, with a slender, tall figure, natural full bust, defined waistline, and an upright, graceful posture. Her temperament is gentle, quiet, high-end, and refined. Long oval elegant face with soft features, gentle willow-leaf eyebrows, slender eyes, natural nose bridge, soft lips, and a thoughtful expression. Black hair half-tied in a low style with side volume, adorned with a green and white jade hairpin and short pearl earrings. Wearing an elegant, semi-sheer [smoky lotus] single-layer crinkle light gauze top over a high-waisted long skirt secured with a narrow silver waistband; the sheer fabric subtly reflects the lake light, showing a natural silhouette while dense wave patterns preserve modesty. She leans sideways against the wooden railing of a lake pavilion, one palm naturally resting on the wood, the other relaxed by her side. Three faint concentric water ripples spread along the lake surface around her, casting subtle spiritual energy. Setting includes the lake pavilion railing, expansive water surface, distant misty mountain rain clouds, and lakeside reeds, illuminated by soft rainy diffuse light.
A textured oil painting in [Renaissance style], depicting a female model filling the frame. Her skin is pale with [a rosy flush and subtle freckles], and her lips are a crimson hue. The background is a textured, greenish-gray stone architectural panel featuring a faint arch motif. Soft, directional lighting emphasizes the rich textures of fabric and skin. Visible canvas texture throughout.
Detailed, text-free animation production character sheet in a hard-edged retro-futurist 1980s/1990s cel-shaded anime style, arranged on a clean flat dark charcoal background in a 16:9 widescreen layout. Characters: - [KES VOSS]: A slim, short-haired young adult with [rusty orange] hair, sharp anime eyes, guarded expression, wearing a worn orange utilitarian pressure suit with black collar, harness straps, rugged boots, patched knees, a mechanical reinforced forearm brace with exposed components and a strapped pressure dial, and an analog chest gauge connected with hoses. - [DARIUS]: A broad, stocky elderly man with dark skin, short white curly hair, white mustache and stubble, heavy brow, weathered face, wearing a bulky blue-gray space survival suit, heavy gloves, suspenders, scuffs, and a torn sleeve patch. Layout and Elements: - Left side ([KES VOSS]): Full-body front turnaround, full-body back turnaround, 3/4 bust portrait, small tense profile looking down, wrist gauge closeup, and boot sole detail. - Right side ([DARIUS]): Full-body front turnaround, full-body back turnaround, 3 head expression studies (including goggle band), collar restraint ring accessory study, and heavy glove detail. - Lower edge: A cinematic horizontal strip of 4 rectangular action panels showing zero-gravity survival scenes—young character welding with bright sparks outside, cockpit/helmet interior closeup, older man repairing machinery amid debris and floating cables, and an exterior shot beside a rugged portable field terminal with a cracked green CRT screen and chunky analog buttons. Style & Atmosphere: Gritty cel-shaded concept art, clean ink linework, muted industrial palette, scuffed metal and fabric, analog space-station grime, dramatic functional lighting. Constraints: Strictly no readable text, no titles, no UI labels, no logos, no watermarks, no chibi proportions.
A side-by-side composition featuring the original person from the reference photo kept fully photorealistic and untouched, standing alongside a [3D caricature companion] version of themselves. The caricature shares the exact identity, outfit, hairstyle, accessories, and facial resemblance, rendered in a [Pixar/Disney-inspired premium 3D styling] with large expressive eyes, a slightly oversized head, smooth cinematic skin, and fashionable proportions standing at [shoulder to ear level] relative to the real person. The real human and the animated caricature interact naturally, [leaning in playfully, smiling, and looking at each other]. Preserve the original background scene, perspective, and lighting from the reference photo, seamlessly integrating the 3D character with realistic cast shadows and cinematic editorial depth.
NovoUm retrato cinematográfico de um personagem com traços faciais nítidos e pele clara e radiante, vestindo um suéter de mohair vermelho, gorro de pele de vison branco e cachecol branco. Parado à noite em uma rua de cidade natalina coberta de neve com luzes penduradas e laços vermelhos, neve caindo suavemente, casas distantes cobertas de neve suavemente desfocadas. Iluminação quente e suave adiciona dimensionalidade, neve no suéter e no cachecol, sorrindo calorosamente, detalhes do rosto e do cachecol nítidos e claros.
NovoTransforme toda a cena, incluindo o fundo e os arredores, em uma cena estática de filme de animação 3D da Pixar de alto orçamento. Os objetos de fundo e o ambiente devem ser renderizados como elementos CGI 3D estilizados com bordas suaves e arredondadas e cores vibrantes de desenho animado, substituindo quaisquer texturas fotorrealistas. O personagem apresenta pele lisa, olhos grandes e expressivos exagerados e proporções fofas em 3D. Renderize tudo usando Unreal Engine 5 com iluminação global cinematográfica brilhante e ray-tracing para garantir que o personagem e o fundo se misturem perfeitamente em um mundo animado em 3D coeso.
Transforme em um estilo de desenho animado bubblegum, mantendo os traços faciais do personagem, paleta de cores doces, contornos grossos, formas suaves e fofas, expressão facial divertida. Se houver conteúdo impróprio (sensível) no fundo, ele pode ser modificado ou removido. Composição redesenhada: mantenha a mesma proporção da imagem original.
Converta a imagem para o estilo de cartum Irasutoya, preservando as feições do rosto do personagem, contornos grossos, blocos de cor pastel chapados, sombras mínimas e expressões fofas. Se houver algum conteúdo inadequado (sensível) no fundo, ele pode ser modificado ou removido. Composição redesenhada: mantenha a mesma proporção da imagem original.
