DeepSeek’s novo modelo está dando trabalho aos concorrentes (literalmente)

[Grego]

The new V3.1 model packs 685B parameters, giving it a larger working memory for longer, more coherent conversations. The new model goes toe-to-toe with OpenAI’s and Anthropic’s flagship offerings while being completely open-source. It landed a score of 71.6% on the prestigious Aider coding benchmark, rivaling Claude Opus 4 at a fraction of the price, costing just $1.01 per coding task compared to nearly $70 for proprietary alternatives.**