North Small Translate

Description

North Small Translate is an open-weights mixture-of-experts (MoE) model purpose-built for machine translation. It has 218 billion total parameters, with 25 billion active parameters, and a 16K context length.

The model provides flexible deployment options for research and enterprise use. Its FP8 weights are available under the Creative Commons Attribution-NonCommercial 4.0 license for non-commercial use. Enterprises can purchase a commercial license and deploy the model through Model Vault.

What can North Small Translate be used for?

North Small Translate is well-suited for:

• Knowledge management: Translate internal documentation, wikis, standard operating procedures, technical manuals, historical reports, and institutional knowledge.
• Safety and operations: Translate safety manuals, emergency procedures, maintenance instructions, manufacturing procedures, incident reports, field-service documentation, and operational alerts.
• Internal communication: Translate company announcements, leadership communications, HR policies, employee handbooks, benefits information, onboarding materials, newsletters, and intranet content.
• Localization and customer support: Translate product content and customer communications while retaining control over where the model and data are hosted.

Supported languages

North Small Translate supports English and more than 50 language and locale variants. Its tier-one languages are Modern Standard Arabic, German, French, Japanese, Korean, Russian, and Ukrainian.

Getting started

North Small Translate is available on the free tier through the Chat V2 API. Ask the model to translate the supplied content into a target language:

1
from cohere import ClientV2
2
3
co = ClientV2(api_key="<YOUR_API_KEY>")
4
5
target_language = "French"
6
content = "Enterprises need accurate translations of business-critical documents."
7
8
response = co.chat(
9
    model="north-small-translate-1-0",
10
    messages=[
11
        {
12
            "role": "user",
13
            "content": f"Translate everything that follows into {target_language}:\n\n{content}",
14
        }
15
    ],
16
)
17
18
print(response.message.content[0].text)

Deployment and licensing

• Cohere API: Try the model on the free tier through the Chat V2 API.
• Open weights: Download the FP8 model from Hugging Face for non-commercial use under the CC BY-NC 4.0 license.
• Commercial deployment: Use a commercially licensed deployment through Model Vault. The suggested hardware is two H100 GPUs or one B200 GPU.

Frequently asked questions

What is machine translation?

Machine translation uses software to convert text from one language into another. A purpose-built generative model can act as the translation engine in an application, workflow, or translation platform.

Why use a machine translation model?

A model gives developers control over integration, prompting, deployment, data residency, and customization. Self-hosted and private deployments can also support sovereignty requirements and reduce dependence on a fixed translation platform.

When is a translation platform a better fit?

A translation platform wraps a model with features such as collaboration, content management, project management, and formatting. Use a platform when those workflow features are more important than direct control over the model.