Google integrates Mistral AIs codestral model: What is it and how will it help developers?
5 months ago | 45 Views
Google Cloud has announced a strategic partnership with the French AI startup Mistral AI, aiming to advance its Vertex AI platform. This collaboration will see the integration of Mistral's innovative code generation model, Codestral, into Google's cloud based AI services.
Mistral AI, known for its advanced technology, has become a prominent player in the European AI landscape. Google Cloud's move to incorporate Codestral represents a significant milestone, positioning it as the first major cloud provider to offer this advanced model as a fully managed service.
In a recent statement as reported by Reuters, Google Cloud highlighted its commitment to open and adaptable AI ecosystems, stating, ?Today, we're introducing Codestral, Mistral AI's first openweight generative AI model explicitly designed for code generation tasks.? This integration aims to equip developers with a powerful tool to automate code writing and enhance development efficiency.
Mistral AI has garnered notable recognition, supported by substantial funding exceeding ?1 billion. The company is now acknowledged as a leading force in the AI sector, attracting significant attention and forming valuable partnerships.
The integration of Mistral AI's Codestral model brings several key benefits:
- Multilingual Capabilities: Mistral AI supports a dozen languages, including English, French, German, Spanish, and Chinese.
- Coding Proficiency: The model is trained on over 80 programming languages, such as Python, Java, C++, and specialised languages like Swift and Fortran.
- AgentCentric Design: Mistral AI offers advanced agentic capabilities, including native function calling and JSON output.
- Advanced Reasoning: It features state-of-the-art mathematical and reasoning skills.
Developers using Google Cloud can access Mistral AI's models by specifying names like mistrallarge@2407 for Mistral Large, mistralnemo@2407 for Mistral Nemo, and codestral@2405 for Codestral. Google recommends using model versions with an @ symbol due to potential version differences.
Mistral AI's rapid growth is further highlighted by recent collaborations, such as its partnership with BNP Paribas to leverage its language models across various business operations.