Scaleway · Tools

Scaleway Generative APIs

Visit website

At a glance

Price
Pricing on request
Vendor
Scaleway

Specifications & properties

Key decision factors

Pricing model
Freemium 1
Provider type
EU provider 1
Data residency
EU only 1
No training on customer data
Yes 1
Model catalogue (LLMs)
Llama 3.3 70B, Mistral, Qwen, Gemma, Whisper and more 1

Pricing

Free tier
Yes 1

Integration

API available
Yes 1
OpenAI-compatible API
Yes 1

Sovereignty & data residency

Data-centre locations
France (Paris) 1

Model

Open-weights models
Yes 1
Report data / suggest a correction

Pay-as-you-go inference API from French cloud provider Scaleway, hosting all models in Paris with a no-read/no-reuse data promise.

Profile

Scaleway Generative APIs

Scaleway Generative APIs is the inference offering of Scaleway, the cloud arm of French group Iliad. Every model runs "in a secure datacenter located in France, Paris only", giving a clear single-country data-residency story that appeals to French and EU regulated buyers.

Scaleway makes a strong privacy commitment: "We do not collect, read, reuse, or analyse the content of your inputs, prompts or outputs generated by the APIs." The catalogue includes open-weight models such as Llama 3.3 70B, Mistral, Qwen and Gemma, all reachable through OpenAI-compatible libraries and LangChain.

Pricing is pay-as-you-go per million tokens, with a free tier of one million tokens to start. Together with OVHcloud, Scaleway is one of the two French sovereign inference platforms, and a natural European alternative to US token APIs.

Frequently asked questions

Where does Scaleway host the models?

All models run in a single secure data centre in Paris, France, giving straightforward single-country data residency.

Is there a free tier?

Yes. Customers receive a free tier of one million tokens, then pay-as-you-go per 1M tokens in/out.