Nano Banana pro 图像生成器
使用先进的推理能力创建令人惊叹的 AI 生成图像
Loading tasks...
什么是 Nano Banana 2?
Nano Banana 2 是一款基于 Nano Banana pro 模型的 AI 图像生成器和编辑器。它专为需要理解视觉指令、执行多步骤编辑并生成逼真高分辨率图像的用户设计。Nano Banana 2 可以同时处理文本和图像,支持混合工作流程,如优化上传的照片、生成变体,或将文本提示与视觉参考相结合。 Nano Banana 2 的独特之处在于它将强大的语义推理与创意灵活性相结合。它能够理解上下文、解析复杂指令,并通过自然语言迭代改进图像——让任何人都能在没有专业技术知识的情况下创作专业级视觉作品。

Nano Banana pro 实战案例
探索创意提示词,看看 Nano Banana pro 能做什么。将鼠标悬停在图片上即可查看提示词并复制它们以用于您自己的创作。

recursive image of an orange cat sitting in an office chair holding up an iPad. On the iPad is the same cat in the same scene holding up the same iPad. Repeated on each iPad.

an office team photo, everyone making a silly face

Create a photo of vagabonds musashi praying drawn on a glass whiteboard in a slightly faded green marker

A hyper-realistic 3D physical object shaped like the [INSERT BRAND NAME] logo, soft studio lighting. The logo is sticker-bombed, covered in a chaotic collage of colorful vibrant stickers.Sticker style: Y2K aesthetic, retro 90s graphics, acid graphics, bold typography, smiley faces, stars, vector badges, "Nano Banana" text, street art vibe, playful branding.Details: Stickers are wrapping around the curves of the logo, slight peeling edges, high-quality texture, tactile feeling.Background: [Isolated on black background OR Blurred natural background].Render: Octane render, 8k resolution, macro photography, depth of field, cinema 4d style, unreal engine 5.

Turn this photo of my [cat/dog] into a funny hand-drawn WeChat sticker . Style : Minimalist ugly-cute line drawing (doodle style). White background. Expression : Exaggerate the animal's expression to look extremely shocked/judgemental/lazy (based on photo). Accessories : Add cute little doodles like sweat drops, question marks, or sparkles around the head. Text : Add handwritten text at the bottom: 'So Dumb'. Ensure the text style is messy and funny.

"Present a clear, 45° top-down view of a vertical (9:16) isometric miniature 3D cartoon scene, highlighting iconic landmarks centered in the composition to showcase precise and delicate modeling.\n\nThe scene features soft, refined textures with realistic PBR materials and gentle, lifelike lighting and shadow effects. Weather elements are creatively integrated into the urban architecture, establishing a dynamic interaction between the city's landscape and atmospheric conditions, creating an immersive weather ambiance.\n\nUse a clean, unified composition with minimalistic aesthetics and a soft, solid-colored background that highlights the main content. The overall visual style is fresh and soothing.\n\nDisplay a prominent weather icon at the top-center, with the date (x-small text) and temperature range (medium text) beneath it. The city name (large text) is positioned directly above the weather icon. The weather information has no background and can subtly overlap with the buildings.\n\nThe text should match the input city's native language.\nPlease retrieve current weather conditions for the specified city before rendering.\n\nCity name:【New York】"

Faithfully restore this image with high fidelity to modern photograph quality, in full color, upscale to 4K

Create a hand drawn isometric schematic diagram image of the building in this photo

"A vibrant Pixar-style 3D animated scene depicting a joyful group selfie moment featuring <group of characters> in a <culturally representative environment>. At the center, <main character> confidently holds a selfie stick topped with an iPhone, wearing an expression that clearly reflects their <distinctive personality trait> and exudes <leadership or core presence>. To the left, <character A> adopts a pose or action reflective of their <distinct personality trait>, showcasing an expressive face that vividly captures their <personality description>. On the right side, <character B> strikes a playful/humorous/cute pose, holding a characteristic item (<character B's representative object>), with an exaggerated, lively facial expression highlighting their <distinctive personality trait>. Additional characters (optional):Nearby, <character C> performs an action or posture aligned with their personality, bearing an expressive facial expression that encapsulates their unique personality traits. All characters wear bright, cheerful, and adorably rounded outfits styled in a contemporary fusion of traditional and modern attire representative of their cultural or historical backgrounds. The scene is warmly lit, colorful, and filled with dynamic expressions and lively poses. The background features a setting emblematic of the characters' cultural identities or personalities—such as cherry blossoms, lakes, mountains, historic architecture, or fantasy-like natural landscapes—rendered in the adorable, cinematic style characteristic of Pixar animations. The overall composition exudes energy, humor, and heartwarming joy, capturing the essence of each character through their selfie expressions and postures.\n\nNames: [Frodo, Sam, Aragorn, Gandalf, Legolas, Gimli]"

Create a 9-image Instagram feed for this product in the same aesthetic. Use different locations, angles, and compositions, incorporating people, animals, nature, and various environments while maintaining a cohesive visual style.

Create a MacBook mockup placed in a minimalist room. Use the attached image on the laptop screen. A woman is sitting at the desk, using the computer, shown from behind.

Create a high-resolution 3D render of attached logo as an inflatable, puffy object. The shape should appear soft, and air-filled, like a plush balloon or blow-up toy. Use a smooth, matte texture with subtle fabric creases and stitching to emphasize the inflatable look. The form should be slightly squishy, with gentle shadows and soft lighting that highlight volume and realism. Place it on a clean, minimal background (light gray)

Create a hand drawn isometric schematic diagram of this street

"The same girl, rendered in an anime-style watercolor touch, is used as a reference image across all scenes to generate cinematic scenes against photorealistic background images.\n\nThe same girl appears throughout, captured in different poses, and expressions. Include artistic close-ups focusing on her hands and feet for subtle realism and poetic atmosphere, and a composition characterized by an impressive foreground blur. Include close-up composition focusing on the character's facial expression. Include overhead shots and low-angle shots."

(full body shot, wide angle view:1.2), [A woman], in a dynamic pose with arm outstretched, wearing a spectacular conceptual avant-garde gown. The bodice is a fitted (colorful patterned patchwork tapestry fabric:1.3). The skirt is NOT fabric and NOT simple ribbons. The skirt is constructed entirely from (GIANT, MASSIVE 3D PERSIAN CALLIGRAPHY STROKES cut from matte paper:1.6). (A violent wind effect blows the calligraphy structure upwards and outwards to the left:1.5), creating a huge swirling anti-gravity vortex of letters that billows like smoke high into the air. The right side of the skirt anchors to the floor. Dramatic color gradient: The floating strokes on the left are (fiery orange and bright red:1.4), transitioning smoothly across the body to (deep teal, emerald green, and dark blue strokes on the bottom right:1.4). White minimalist studio background with (distinctive patterned gobo shadows cast on the wall:1.3). Cinematic lighting, hyper-detailed paper texture, masterpiece, 8k. (simple ribbons with writing on them:1.6), (thin paper strips:1.5), (small text:1.4), standard fabric skirt, normal dress, metallic texture, shiny plastic, glossy finish, wire mesh, angel wings, feathers attached to back, symmetrical skirt, messy background, low quality, blurry.

"画中描绘着一位绝美的卡通风女子正面特写头像,她神态柔美而宁静。墙体顶部被一大片盛开的蔷薇花覆盖,茂密的绿叶与繁盛的花朵向外舒展,部分枝条从墙顶垂落而下,与女子的头发巧妙融合,使她的秀发宛如由层层叠叠的蔷薇花组成。这些繁密的花朵簇拥着女子的头部,形成了一顶瑰丽的花冠,视觉效果华美浪漫。\n\n背景中蓝天澄澈,点缀着朵朵白云;地面为一条细节真实的沥青街道,上面散落着缤纷多彩的花瓣,行人悠然漫步其间。整体场景细节精致入微,光影明亮柔和,营造出犹如现实般的梦幻街景氛围。"

create a sprite sheet for a 2D NES platforming game, include enemies and NPCs
Nano Banana 2 核心功能
高分辨率图像生成(1K / 2K / 4K)
Nano Banana 2 支持最高 4K 分辨率输出,为商业设计、广告、产品渲染和印刷级视觉提供清晰锐利的效果。

多轮图像编辑
Nano Banana 2 支持无缝迭代编辑。用户可以逐步优化细节。模型会记住跨轮次的上下文,提供稳定一致的编辑效果。

文本-图像混合生成
Nano Banana 2 支持文本和图像混合输入,实现直观的工作流程,如优化参考照片、融合草图、合并多张图像,以及通过简单的文字指令调整风格或布局。

可配置宽高比
选择任意宽高比——1:1、16:9、4:5、电影级、竖屏、超宽屏等。Nano Banana 2 自动适配构图,为创作者在社交媒体、广告、网站和印刷品等平台提供自由度。

增强角色一致性
与 Nano Banana 1 相比,Nano Banana 2 在多张图像中提供更强的面部和身份一致性,保持面部特征、表情、服装逻辑和比例对齐——非常适合故事叙述和品牌视觉。

多图融合
用户可以上传多张照片,让 Nano Banana 2 将它们融合成一个连贯的场景。光照、阴影、透视和风格会自动协调统一。

Google 搜索增强
Nano Banana 2 可以选择性地使用 Google 搜索增强来提高事实准确性——对产品设计、现实世界物体或参考感知创作非常有帮助。

Nano Banana 2 常见问题
Nano Banana 2 与 Nano Banana 1 有什么区别?
Nano Banana 2 新增了 4K 输出、多轮编辑、更稳定的角色一致性、更强的推理能力、批量生成和可配置宽高比。
使用 Nano Banana 2 需要设计经验吗?
不需要。Nano Banana 2 接受自然语言指令,并为您处理复杂的编辑任务。
Nano Banana 2 可以编辑上传的照片吗?
可以。它可以修图、增强、修改场景,或合并多张上传的图像。
Nano Banana 2 支持哪些分辨率?
支持 1K、2K 和 4K 图像生成。
