Scaleway Generative APIs

Get 1 million free tokens to access DeepSeek, Llama, Gemma, Qwen, and more

Overview

Scaleway Generative APIs provides 1 million free tokens to new users, giving you access to a diverse selection of open-source language models, embedding models, and audio models.

Trial Tokens

1,000,000 free tokens

Model Variety

20+ models available

Available Models

Scaleway offers a comprehensive selection of models across different categories:

Language Models

Model	Description
DeepSeek R1 Distill Llama 70B	Distilled reasoning model
Gemma 3 27B Instruct	Google’s efficient model
Llama 3.1 8B Instruct	Fast Llama model
Llama 3.3 70B Instruct	Large Llama model
Mistral Nemo 2407	Mistral’s efficient model
mistral-small-3.2-24b-instruct-2506	Latest Mistral Small
devstral-2-123b-instruct-2512	Large development model
gpt-oss-120b	Open-source GPT model
holo2-30b-a3b	Specialized model
qwen3-235b-a22b-instruct-2507	Large Qwen model
qwen3-coder-30b-a3b-instruct	Coding-focused Qwen

Vision Models

• Pixtral 12B (2409): Multimodal vision-language model

Audio Models

• Whisper Large v3: State-of-the-art speech recognition
• voxtral-small-24b-2507: Voice processing model

Embedding Models

• BGE-Multilingual-Gemma2: Multilingual embeddings
• qwen3-embedding-8b: Qwen embeddings

Getting Started

1. Sign Up

Visit console.scaleway.com/generative-api/models and create a Scaleway account.

2. Get Your API Key

Navigate to the Generative API section in the console to generate an API key.

3. Make API Calls

import openai

client = openai.OpenAI(
    api_key="YOUR_SCALEWAY_API_KEY",
    base_url="https://api.scaleway.ai/v1"
)

response = client.chat.completions.create(
    model="llama-3.3-70b-instruct",
    messages=[{
        "role": "user",
        "content": "What is the capital of France?"
    }]
)

print(response.choices[0].message.content)

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.SCALEWAY_API_KEY,
  baseURL: 'https://api.scaleway.ai/v1'
});

const response = await client.chat.completions.create({
  model: 'llama-3.3-70b-instruct',
  messages: [{
    role: 'user',
    content: 'What is the capital of France?'
  }]
});

console.log(response.choices[0].message.content);

curl https://api.scaleway.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_SCALEWAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "llama-3.3-70b-instruct",
    "messages": [{
      "role": "user",
      "content": "What is the capital of France?"
    }]
  }'

Model Categories

Text Generation

DeepSeek, Llama, Gemma, Qwen, Mistral models

Code Generation

Qwen Coder, Devstral for development

Embeddings

BGE and Qwen embeddings for RAG

Multimodal

Pixtral for vision-language tasks

Use Cases

Text & Code Generation

• Chatbots: Build conversational AI with Llama or Qwen
• Code Assistance: Use Qwen Coder or Devstral for coding
• Content Creation: Generate text with various models

Embeddings & RAG

• Semantic Search: Use embedding models for similarity
• RAG Systems: Build retrieval-augmented generation apps
• Document Analysis: Process and understand documents

Audio & Vision

• Transcription: Convert speech to text with Whisper
• Image Understanding: Analyze images with Pixtral

European Infrastructure

Scaleway is a European cloud provider with data centers in Europe, offering GDPR-compliant infrastructure for AI workloads.

Benefits

• GDPR Compliance: European data residency
• Low Latency: Fast access from Europe
• Reliable Infrastructure: Enterprise-grade cloud platform

API Compatibility

Scaleway uses an OpenAI-compatible API, making it easy to integrate with existing applications and tools.

Resources

Scaleway Console

Access the platform

Documentation

View API documentation

With 1 million free tokens, you have substantial capacity to test and evaluate multiple models for your use case.