(Reuters) -Artificial intelligence startup Anthropic unveiled an upgraded Opus model on Monday, boosting Claude's ability to write detailed code, create sophisticated agents and streamline enterprise workflows through spreadsheet and financial analysis.

The new model comes as Amazon ( AMZN ) and Alphabet-backed Anthropic races against OpenAI and other rivals to develop cutting-edge large language models aimed at achieving capabilities that could surpass human intelligence.

Opus 4.5 ranks among the most powerful models in the Claude family, offering deep reasoning and memory, coding and a versatile performance across a range of computer applications, including financial tasks such as modeling and forecasting.

Its agents autonomously refine their own capabilities and store insights from past work to apply at a later date, Anthropic said.