Claude 2 - Anthropic’s best publicly available model (11 jul 2023 ano – 3 mar 2024 ano)

Descrição:

Anthropic, an AI startup, released Claude 2, a large-language model that is designed to be more capable, less gullible, and less harmful than its predecessor. Claude 2 can handle more complex prompts, is less likely to generate harmful or offensive responses, and is more resistant to "jailbreak" prompts. It also scores in the 90th percentile of graduate school applicants on the GRE reading and writing exams.

Adicionado na linha do tempo:

Data: