Meta open-sourced its large language model LLaMA 2, positioning it against OpenAI's GPT-4, with the announcement made during Microsoft's Inspire event. The model is said to be trained on 40% more data than LLaMA 1 and will be accessible through Microsoft's Azure, with plans for other providers.
One outlet so far Only Maldives Republic has reported this among the sources we monitor. A gap in what we monitor, not proof that others did not report it.
Nothing to compare yet. If another monitored source reports it, its headline appears beside this one.
The LLaMA 2 model has been trained on 40% more data compared to its predecessor, LLaMA 1, and is said to outperform other language models like Falcon and MPT in various areas, including reasoning, coding, proficiency, and knowledge tests.
Word for word from the report in Maldives Republic. Open the report
The previous model already received more than 100,000 requests from researchers, indicating that the open-source LLaMA 2 will have an even broader reach and impact in the AI community.
Word for word from the report in Maldives Republic. Open the report