While AI companies trade blows for the top spot as new updates are released, what makes DeepSeek V3's accomplishment noteworthy is that it's the first time an open-weights model is leading the pack.

China’s fast-growing AI scene had a big moment this week as Beijing-based startup Z.ai, formerly known as Zhipu, rolled out ...

Startup Z.ai, formerly known as Zhipu, announced Monday a new, low-cost AI model as Chinese companies race to stay at the ...

Despite the restrictions, Chinese firms kept training world-beating AI models—Kimi K 2, unveiled in July by Moonshot AI, a ...

DeepSeek has seen use of its models drop to 80 per cent from 99 per cent on Chinese cloud computing platform PPIO.

Benchmark tests show DeepSeek-V3 achieves 90% accuracy in mathematical problem-solving. This precision supports complex tasks such as advanced calculations, data analysis, and logical reasoning.

It has also rolled out specially distilled 7B and 14B versions optimized for Copilot+ PCs through Azure AI Foundry. And recently, Microsoft incorporated DeepSeek-V3 into Azure's model catalog.

Deepseek has just launched its latest model, DEEPSEEK-V3-0324, made available via the AI development platform Hugging Face (this does require some coding knowledge to try out for now).

Last, DeepSeek’s achievements appear to be a validation that open approaches to AI development may accelerate those efforts. 17 By releasing detailed research reports and providing the models ...