# How Krisp uses Nebius AI Cloud

**50–80%**: “a single H100 GPU completed the training process 50–80% faster than the A100 GPU”

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

- Company: [Krisp](https://www.usedby.ai/companies/krisp.md)
- Tool: [Nebius AI Cloud](https://www.usedby.ai/tools/nebius.md)
- Industry: [Artificial Intelligence](https://www.usedby.ai/companies/industry/artificial-intelligence.md)
- Teams: research, engineering

## What the story says

Krisp trains its large foundational voice conversion models for Accent Localization on Nebius H100 GPUs, which are later used to build compact real-time models that run on-device. The work also uses Nebius high-speed direct attached SSDs for data handling during training.

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

> They found that a single H100 GPU completed the training process 50–80% faster than the A100 GPU they were familiar with.

## What usedby checked

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

- Checked: the figure 50–80% is printed word for word on the page, near the name of Krisp.
- Checked: the passage quoted above is copied word for word from the page, near the name of Krisp.
- Checked: each number in our summary is printed on the page.
- Checked: Krisp 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/krisp-nebius · How we check: https://www.usedby.ai/methodology