Google 图片 生成器
Google Nano Banana AI, also called Gemini 2.5 Flash Image, is an advanced AI image generation and editing model from Google DeepMind. It is designed for detail-rich image creation, highly consistent iterative edits, and fast natural-language-driven refinement across both original generation and image editing workflows.
Google 的功能亮点
汇总该提供方主要模型系列中的共性优势。
State-of-the-art Image Generation
Google Nano Banana AI combines strong prompt understanding with contextual consistency, making it well-suited for producing high-quality images that stay aligned with detailed text instructions across a wide range of art directions.
Prompt
A digital illustration of a glowing forest spirit sitting beside bioluminescent flowers in a misty valley at sunrise, cinematic composition, soft painterly textures, high detail.
输出图片
Prompt
A young Asian female figure, positioned in the center of the image. She appears to be a teenager, with long, straight, dark black hair styled in pigtails. She is wearing a light beige-colored cardigan-style sweater over a white collared shirt, a plaid mini skirt, black socks, and brown oxfords.
输出图片
| Prompt | 输出图片 |
|---|---|
A digital illustration of a glowing forest spirit sitting beside bioluminescent flowers in a misty valley at sunrise, cinematic composition, soft painterly textures, high detail. | |
A young Asian female figure, positioned in the center of the image. She appears to be a teenager, with long, straight, dark black hair styled in pigtails. She is wearing a light beige-colored cardigan-style sweater over a white collared shirt, a plaid mini skirt, black socks, and brown oxfords. | |
Perfect Character Consistency
Nano Banana is designed for revision-heavy workflows where you need to add, remove, or adjust scene details without losing subject identity, composition, or overall style consistency from one edit to the next.
Original Image
Original Image 2
Prompt
Place the subject into a cozy window-side cafe in Paris. Keep the red scarf, facial features, and illustration style consistent while updating the environment lighting and background details.
Output Image
| Original Image | Original Image 2 | Prompt | Output Image |
|---|---|---|---|
| | Place the subject into a cozy window-side cafe in Paris. Keep the red scarf, facial features, and illustration style consistent while updating the environment lighting and background details. | |
Natural Language Image Refinement
Using simple text instructions, Nano Banana can identify edit regions and make focused changes such as swapping backgrounds, removing distractions, changing poses, or refining stylistic details with minimal overhead.
Original Image
Prompt
Remove the background crowd, add soft golden-hour sunlight, and update the outfit to a light linen jacket while keeping the subject framing and expression consistent.
Output Image
| Original Image | Prompt | Output Image |
|---|---|---|
| Remove the background crowd, add soft golden-hour sunlight, and update the outfit to a light linen jacket while keeping the subject framing and expression consistent. | |
Seamless Multi-Image Fusion
Gemini 2.5 Flash Image effortlessly understands and blends multiple input images. Whether inserting an object into a scene, restyling a room with new colors or textures, or merging images, you can achieve stunning results with a single natural language prompt.
Ultra-Fast Image Processing
Compared to other AI image models, Google Nano Banana AI takes speed to a whole new level. It doesn’t process imagery in tens of seconds but in even less time than that. Because of this, it can be an ideal solution for rapid ideation, experimentation, and editing, making it easier than ever for creators to produce more refined projects.
如何在 skills.video 中使用 Google
01
选择一个 Google 模型
先挑选最适合你工作流的模型系列，再进入对应的模型详情页。
02
查看示例与功能
在生成前先检查每个模型的示例、控制项和功能细节。
03
开始创作
选好模型后，直接进入图片或视频生成流程。
图片模型
对比 Google 的图片模型，适用于文生图与图像编辑工作流。
常见问题
关于 Google 模型和工作流的常见问题。