Posts

Showing posts with the label DeepSeek

A Brief Review on DeepSeek to Know What's It Is !!

Image
DeepSeek is a Chinese artificial intelligence (AI) company based in Hangzhou, Zhejiang, founded in 2023 by CEO Liang Wenfeng. company has rapidly gained attention for its open-source large language models (LLMs), notably the DeepSeek series, which have been recognized for their efficiency and performance. ( .wikipedia.org ) Technological Innovations DeepSeek's models, such as DeepSeek-V2 and DeepSeek-r-V2, have introduced significant advancements in AI. DeepSeek-V2 employs a Mixture-of-Experts (MoE) architecture, activating only a subset of its parameters during inference, leading to cient computation and reduced resource consumption. This model supports a context length of up to 128,000 tokens, falitating complex taskseiring extensive context. ( arxiv.org ) DeepSeek-Coder-V2, an extension of DeepSeek-V2, focuses on code intelligence. It supports 338 programming languages and nds the context length to 128,000 tokens, enhancing its applicability in code-related tasks....