DeepSeek is calling these new models DeepSeek-GRM — short for “generalist reward modeling” — and will release them on an open source basis, the company said.
DeepSeek is working with Tsinghua University on reducing the training its AI models need in an effort to lower operational costs.
Reinforcement learning has proven effective in speeding up AI tasks in narrow applications and spheres. The strategy outperformed existing methods and models on various benchmarks and the result showed better performance with fewer computing resources, according to the paper.
South Africa Latest News, South Africa Headlines
Similar News:You can also read news stories similar to this one that we have collected from other news sources.
Interest in DeepSeek is fodder for cybercriminalsA fake app that claims to be the recently launched DeepSeek AI, turns out to be nothing but a bit of malware.
Read more »
Microsoft working on its own AI in spite of ChatGPT investmentsMicrosoft is reportedly working on its own in-house generative AI models and is planning to release them to customers.
Read more »
OpenAI improves real-time conversations for ChatGPTOpenAI has announced an update to its Advanced Voice Mode for ChatGPT that includes less interruptions and sounding more natural.
Read more »
Does "vibe coding" make everyone a programmer?Can a complete tech novice create a website using everyday language on ChatGPT?
Read more »
OpenAI study finds links between ChatGPT use and lonelinessHigher use of chatbots like ChatGPT may correspond with increased loneliness, according to new research.
Read more »
ChatGPT can now generate more coherent text in imagesWith GPT-4o, ChatGPT users can now make images with text you can read.
Read more »