SEEDANCE 2.0 & SEEDANCE 2.0 FAST UP TO 36% OFF

skills.video logo
Z.ai提供方概览

Z-Image 图片 生成器

Z-Image 是 Tongyi-MAI 的图像模型系列，包含生成与编辑等变体，强调高质量输出、良好审美表现，以及可控的提示词驱动创作能力。

探索 Z-Image 的模型

直接进入你想比较、测试或用于生成的具体模型页面。

Z-Image Turbo

Z-Image 的功能亮点

汇总该提供方主要模型系列中的共性优势。

Text-To-Image GenerationImage-To-Image GenerationCompact Sampling ControlsSize And Batch Options

Z-Image Turbo

Text-To-Image Generation

When no input image is provided, the connector calls fal-ai/z-image/turbo. The model is designed for very fast prompt-driven image generation and defaults to a landscape_4_3 image size on fal.

Z-Image Turbo

Image-To-Image Generation

When image_url is provided, the connector calls fal-ai/z-image/turbo/image-to-image. This mode adds image_url as a required upstream parameter and supports strength to control how strongly the input image conditions the output.

Z-Image Turbo

Compact Sampling Controls

Z-Image Turbo is optimized for a short sampling path. The fal schema supports num_inference_steps from 1 to 8, plus an optional seed for reproducible outputs with the same prompt and model version.

Z-Image Turbo

Size And Batch Options

The text endpoint supports the standard fal image-size presets, while image-to-image also supports auto sizing. Both endpoints expose num_images with a 1 to 4 range.

如何在 skills.video 中使用 Z-Image

01

选择一个 Z-Image 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 Z-Image 的图片模型，适用于文生图与图像编辑工作流。

1 个模型
Z.ai

Z-Image Turbo

Tongyi-MAI 的快速 6B Z-Image Turbo 模型，支持文生图与图生图。

常见问题

关于 Z-Image 模型和工作流的常见问题。

What is Z-Image Turbo?expand_more
Z-Image Turbo is a fast 6B-parameter image generation model developed by Tongyi-MAI and exposed on fal for low-latency image generation.
Does Z-Image Turbo support text-to-image?expand_more
Yes. Text-to-image requests use the fal-ai/z-image/turbo endpoint and require only a prompt.
Does Z-Image Turbo support image-to-image?expand_more
Yes. Image-to-image requests use the fal-ai/z-image/turbo/image-to-image endpoint and require both prompt and image_url. The strength parameter controls conditioning strength.
Which image sizes are supported?expand_more
The model supports fal image-size presets such as square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3, and landscape_16_9. Image-to-image also supports auto.
Which output formats are supported?expand_more
Z-Image Turbo supports jpeg, png, and webp output formats.
How many images can one request generate?expand_more
The fal schema supports num_images from 1 to 4.
skills.video