{"id":7265,"date":"2024-11-18T18:51:18","date_gmt":"2024-11-18T16:51:18","guid":{"rendered":"https:\/\/blog.outscale.com\/?p=7265"},"modified":"2025-09-03T10:36:55","modified_gmt":"2025-09-03T08:36:55","slug":"how-to-deploy-codestral-on-outscales-sovereign-and-secure-infrastructure-2","status":"publish","type":"post","link":"https:\/\/blog.outscale.com\/en\/how-to-deploy-codestral-on-outscales-sovereign-and-secure-infrastructure-2\/","title":{"rendered":"How to deploy Codestral on OUTSCALE’s sovereign and secure infrastructure?"},"content":{"rendered":"

Generative AI<\/strong> is playing an increasingly significant role in software development<\/strong>. By generating precise code snippets, automating testing, and simplifying documentation, these LLMs (Large Language Models) enhance developers’ productivity while freeing up time for innovation.<\/p>\n

However, the rise of Generative AI raises a critical question: how can we ensure the sovereignty and security of the generated code?<\/strong><\/p>\n

LLM and Asset Security<\/h2>\n

Companies and institutions regularly handle sensitive data and codes, but using AI tools hosted on online platforms may expose these strategic assets to various risks.<\/p>\n

First, extraterritorial laws like the Cloud Act<\/strong> in the United States allow certain governments to request access to data stored within their jurisdictions, potentially compromising the confidentiality of critical information.<\/p>\n

Second, industrial espionage<\/strong> and cyberattacks increase the risk of intrusion, code theft, or even sabotage, particularly through supply chain attacks<\/strong>.<\/p>\n

To address these threats, organizations must consider solutions that combine the power of Generative AI<\/strong> with guarantees of security<\/strong>, data sovereignty<\/strong>, and intellectual property protection<\/strong>.<\/p>\n

Dassault Syst\u00e8mes and Mistral AI Partnership<\/h2>\n

In July 2024, Dassault Syst\u00e8mes and Mistral AI announced a strategic partnership to deliver reliable industrial solutions based on artificial intelligence to accelerate the Generative Economy<\/a>. This partnership combines the virtual twin experiences and the sovereign Cloud infrastructure of OUTSCALE (Dassault Syst\u00e8mes) with the large language models (LLMs) from Mistral AI<\/a>.<\/p>\n

In this context, OUTSCALE has launched an LLM as a Service (LLMaaS) offering<\/strong>, specifically designed to enable rapid development of Generative AI use cases<\/strong>.<\/p>\n

This offering is the first to integrate Mistral AI models on a 100% sovereign<\/strong>, secure Cloud qualified with SecNumCloud 3.2<\/strong>, ensuring top-tier performance while providing enhanced protection for sensitive data and intellectual property.<\/p>\n

Available Mistral AI Models<\/h2>\n

 <\/p>\n

\n

\"Available<\/p>\n

 <\/p>\n

As part of their partnership, OUTSCALE and Mistral AI offer a range of Artificial Intelligence models designed to meet the specific needs of companies and institutions.<\/p>\n

These models include Mistral Small<\/strong><\/a> and Codestral<\/strong><\/a>, integrated into the LLMaaS (Large Language Model as a Service) deployed on OUTSCALE\u2019s sovereign Cloud infrastructure.<\/p>\n

This offer allows organizations to fully leverage the capabilities of Generative AI, in a sovereign and secure environment.<\/p>\n

\"\"<\/p>\n

Mistral Small: A Lightweight and Versatile Model<\/h3>\n

Mistral Small<\/strong> is designed to offer lightweight Generative AI solutions while optimizing resource consumption. Ideal for projects requiring speed and efficiency without compromising quality, it adapts to a wide range of tasks such as text generation<\/strong>, auto-completion<\/strong>, and data processing<\/strong> from simple instructions. Thanks to its reduced latency, Mistral Small is particularly well-suited for real-time applications or environments that require quick responses.<\/p>\n

Mistral Small<\/strong> is effective for:<\/p>\n