# Alibaba Releases Z-Image-Turbo, A Fast Open Image Model

> The new text-to-image model from the team behind Qwen uses a diffusion transformer to generate high-resolution images in just a single step.

Published by The Open Weights on Nov 25, 2025. Canonical: https://theopenweights.com/news/z-image-turbo-kk9g

## Key facts

- Company: Qwen · Alibaba
- Model: Z-Image-Turbo
- Version: Turbo
- Category: Text → Image
- Modalities: Text → Image
- License: Apache 2.0 (Open weights, commercial use allowed)
- Architecture: ZImagePipeline
- Precision: FP32
- Weights size: 24.6 GB
- Significance: notable
- Published: 2025-11-25
- Last verified: 2026-06-17
- Hugging Face: https://huggingface.co/Tongyi-MAI/Z-Image-Turbo
- Canonical URL: https://theopenweights.com/news/z-image-turbo-kk9g

Alibaba's Tongyi-MAI team, the creators of the Qwen model family, has released Z-Image-Turbo, a new open-source text-to-image model designed for speed. The model stands out for its ability to generate a 1024x1024 pixel image in a single inference step, a significant departure from the multi-step sampling process required by most diffusion models.

This efficiency is achieved by using a Diffusion Transformer (DiT) architecture. This approach, which has gained prominence in models like Sora, moves away from the U-Net architectures common in earlier Stable Diffusion versions. By building on a transformer backbone, Z-Image-Turbo can capture complex relationships in text prompts more effectively and streamline the generation process.

The combination of speed and a modern architecture makes Z-Image-Turbo a compelling new option for developers. The model and its weights are available on [Hugging Face](https://huggingface.co/Tongyi-MAI/Z-Image-Turbo) under the permissive Apache 2.0 license, which allows for commercial use.

For the open-source AI ecosystem, Z-Image-Turbo represents another powerful, commercially-friendly alternative to established models. Its single-step generation offers a practical advantage for applications where low latency is critical, such as real-time previews or interactive image creation tools.

## Get the model

- [Hugging Face](https://huggingface.co/Tongyi-MAI/Z-Image-Turbo)
- [civitai](https://civitai.com/models/2168935)

## Sources

- [Tongyi-MAI/Z-Image-Turbo](https://huggingface.co/Tongyi-MAI/Z-Image-Turbo) — Hugging Face, Nov 25, 2025
- [Z Image Turbo](https://civitai.com/models/2168935) — civitai, Nov 26, 2025

---
Source: The Open Weights (https://theopenweights.com/). Aggregated and written by Claude, curated by humans. Cite as: "Alibaba Releases Z-Image-Turbo, A Fast Open Image Model", The Open Weights, Nov 25, 2025, https://theopenweights.com/news/z-image-turbo-kk9g