Meta is pitching its AI technology to beat ChatGPT

meta announced that it is open-sourcing its large language model llama 2Making it free for commercial and research use and going head-to-head with OpenAI’s free-to-use GPT-4, which powers tools like ChatGPT and Microsoft Bing.

Meta announced the move As part of Microsoft’s Inspire event, noting its support for Azure and Windows and the “growing” partnership between the two companies. Plus, Microsoft revealed more details about the AI ​​tools built into its 360 platform and how much they’ll cost. Qualcomm also announced It is working with META to bring LLAMA to laptops, phones and headsets starting in 2024 for AI-powered apps that work without relying on cloud services.

Meta’s press release explains the decision to open LLAMA as a way to give businesses, startups and researchers access to more AI tools, allowing experimentation as a community. according to meta, LLaMa 2 was trained on 40 percent more data than LLaMa 1, including information from “publicly available online data sources”. It also states that it “outperforms” other LLMs like Falcon and MPT when it comes to reasoning, coding, proficiency and knowledge tests.

“We believe an open approach is right for the development of today’s AI models, especially in the generative sector where technology is advancing rapidly,” Meta said in a statement. “Opening up access to today’s AI models means that a generation of developers and researchers can test them as a community, identify problems, and solve them faster.”

The open-sourced LLaMa 2 will be available through Microsoft’s Azure platform. Meta said LLaMA will also be available through AWS, Hugging Face and other providers.

While Meta first announced its LLaMA model in February, it leaked onto 4chan just days later. Opening the LLaMA, Meta said it is looking to improve security and transparency. The company said the LLaMA 2 model has been “red-teamed”, or tested for safety internally and externally “by generating adversarial signals to facilitate model fine-tuning”. The meta also describes how models are evaluated and modified.