# How Recraft uses Nebius AI Cloud

**~6x**: “These adjustments made the network 4x faster, leading to a ~6x increase in training speed.”

As published on [nebius.com](https://nebius.com/customer-stories/recraft). Captured by usedby on 2026-10-07.

- Company: [Recraft](https://www.usedby.ai/companies/recraft.md)
- Tool: [Nebius AI Cloud](https://www.usedby.ai/tools/nebius.md)

## What the story says

Recraft trained its text-to-image foundation model from scratch on Nebius, using Managed Kubernetes with Kubeflow and PyTorch for distributed GPU training. Nebius engineers and solution architects helped resolve network and hardware issues during training.

Summary written by usedby from the source page, in English. The figures are those of Nebius AI Cloud and Recraft, not ours.

> After the initial problems were solved, Recraft noticed the training process was very slow, operating at an 8x slower rate than in an ideal scenario where the network is not a bottleneck.

> Nebius stands out when compared to other clouds, especially if you’re looking for flexibility and quick support. On the scalability front, they always have GPUs available, so no frustrating delays waiting for quotas. The storage speed on Nebius is also higher in comparison to some other clouds. Nebius offers more control, better support and reliable scalability compared to some other clouds.
>
> Anna Veronika Doroqush, Founder and CEO

> If you have simple code, you can switch to the new GPUs without doing anything special,
>
> Pavel Ostyakov, Head of AI at Recraft

## What usedby checked

We compared the story with its live page on 2026-10-07.

- Checked: the figure ~6x is printed word for word on the page, near the name of Recraft.
- Checked: the passage quoted above is copied word for word from the page, near the name of Recraft.
- Checked: Recraft uses Nebius AI Cloud. Confirmed line. Latest check across sources: 2026-10-07.
- Not checked: the result itself. We quote it; we did not measure it.

---
Source: https://www.usedby.ai/case-studies/recraft-nebius-2 · How we check: https://www.usedby.ai/methodology