What you need to know
- Microsoft and Meta are extending their long-standing partnership.
- Meta has open-sourced Llama 2, allowing more developers to leverage its capabilities.
- Llama 2 is now supported on Azure and Windows.
- The tool will also be available across AWS, Hugging Face, and more.
Microsoft recently held its annual Inspire event, making several announcements, including the cost for Microsoft 365 Copilot, Visual Search in Bing Chat, and Bing Chat Enterprise. Aside from these developments, the company also announced its extended partnership with Meta.
With the extension of this long-standing partnership, the Llama 2 family of large language models (LLMs) will now be supported on Azure and Windows. For those unfamiliar, Llama 2 is a tool that's designed to assist developers in building AI-powered tools.
However, AI models haven't been openly accessible to everyone. But with Meta now open-sourcing Llama 2, more developers will be able to leverage its capabilities to venture into AI and possibly tap into a whole new dimension. Access to such facilities has been limited, considering the magnitude of the computing power.
Microsoft explained the benefits of the partnership:
Now Azure customers can fine-tune and deploy the 7B, 13B, and 70B-parameter Llama 2 models easily and more safely on Azure, the platform for the most widely adopted frontier and open models. In addition, Llama will be optimized to run locally on Windows. Windows developers will be able to use Llama by targeting the DirectML execution provider through the ONNX Runtime, allowing a seamless workflow as they bring generative AI experiences to their applications.
By open-sourcing Llama 2, Meta can now compete on an even playing field with other key players in the industry, including OpenAI's ChatGPT and more. Meta further indicated, "We believe an open approach is the right one for the development of today’s AI models, especially those in the generative space where the technology is rapidly advancing. By making AI models available openly, they can benefit everyone."
Now, more businesses and organizations can access more AI tools via Llama 2, thus creating more room for creation and experimentation. What's more, compared to its previous iteration, Llama 1, Llama 2 is more advanced because it was trained using 40 percent more data, including data from public online sources. It also stands out from similar models like Falcon and MPT while performing tasks such as coding, reasoning, and more.
Meta has also indicated that it has taken extra precautions to ensure that the platform is secure despite it being open-sourced. It was able to achieve this by “generating adversarial prompts to facilitate model fine-tuning.”
Microsoft has also indicated that Meta selecting Azure as its preferred " strategic cloud provider" lines up perfectly. This is because "Azure’s purpose-built AI supercomputing platform is uniquely designed from the facility, hardware and software to support the world’s leading AI organizations to build, train and deploy some of the most demanding AI workloads."
Additionally, Microsoft noted that the incorporation of Llama 2 models into Windows makes the platform an ideal place for developers to build AI experiences that are designed to meet the needs of the users since they'll also have access to essential tools such as Windows Subsystem for Linux (WSL), Windows terminal, Microsoft Visual Studio, and VS Code.
Besides Azure, Meta indicated that Llama 2 will be available across Amazon Web Services, Hugging Face, and more.
