ETtech Explainer: The story behind DeepSeek’s greener and leaner chatbot
Chinese AI startup DeepSeek’s cost-efficient technology propelled it to overnight fame leaving industry insiders, investors and competitors stunned by the sudden disruption. Here’s a look at the underlying technology.
Budget with ET
India, get ready for the modern warfare of tomorrow
Women, youth, farmers and poor can continue to be Budget 2025's ‘roti, kapda aur makan’
Modi govt has a key task in Budget 2025: Unlocking the PLI goldmine
How does DeepSeek's tech work?
DeepSeek V3, the large language model (LLM) that operates the chatbot R1, matchesl or outperforms competitors like OpenAI’s ChatGPT 4.o1 on various benchmarks. The system employs a machine-learning technique known as the “Mixture of Experts” (MoE) architecture.
This method utilises multiple specialised models, termed “experts,” to address various aspects of a task. Each expert is trained in a specific domain. DeepSeek’s architecture activates only 37 parameters out of 671 billion for each task, a remarkable feat that significantly enhances computational capacity while reducing costs.
Also Read: ETtech Explainer: What is DeepSeek, China's competitor to OpenAI?
Artificial Intelligence(AI)
Java Programming with ChatGPT: Learn using Generative AI
By — Metla Sudha Sekhar, IT Specialist and Developer
Artificial Intelligence(AI)
Basics of Generative AI: Unveiling Tomorrows Innovations
By — Metla Sudha Sekhar, IT Specialist and Developer
Artificial Intelligence(AI)
Generative AI for Dynamic Java Web Applications with ChatGPT
By — Metla Sudha Sekhar, IT Specialist
