Sources say that Meta will release a commercial version of the LLaMA model.

CTOnews.com, July 13 (Xinhua)-- Meta is preparing to release a commercial version of its artificial intelligence model, allowing startups and companies to build custom software based on the technology, according to the Financial Times. The move will put Meta in competition with Microsoft-backed OpenAI and Google.

Earlier this year, Meta released its own large language model LLaMA to researchers and scholars. The new version will be more widely available to companies and can be customized by companies, according to three people familiar with the matter.

According to previous reports from CTOnews.com, Meta released its own large language model LLaMA in February this year, with parameters ranging from 7 billion to 65 billion. Even LLaMA (13 billion), which only uses 10 parameters, can surpass GPT-3; in most benchmarks, while LLaMA with 65 billion parameters is on a par with DeepMind's Chinchilla (70 billion parameters) and Google's PaLM (540 billion parameters).

Although Meta advertised it as open source, someone leaked the LLaMA model file just days after its release. In just a few weeks, the innovation of LLM agents built on it has exploded, such as Alpaca, Vicuna, Koala, ChatLLaMA, FreedomGPT, ColossalChat …

Related readings:

"Meta launched the AI language model LLaMA, a large language model with 65 billion parameters."