DeepSeek releases model it calls 'intermediate step' towards 'next-generation architecture'


The logo of DeepSeek is displayed alongside its AI assistant app on a mobile phone, in this illustration picture taken January 28, 2025. REUTERS/Florence Lo/Illustration

BEIJING (Reuters) -Chinese AI developer DeepSeek has released its latest model which it said was an "experimental release" that was more efficient to train and better at processing long sequences of text than previous iterations.

The Hangzhou-based company called DeepSeek-V3.2-Exp an "intermediate step toward our next-generation architecture" in a post on developer forum Hugging Face.

Follow us on our official WhatsApp channel for breaking news alerts and key updates!

Next In Tech News

Tech tracking to tackle human-wildlife conflict in Zimbabwe
Like fancy Japanese toilets? You’ll love the sound of this.
Facebook 'supreme court' admits 'frustrations' in five years of work
Russia restricts FaceTime, its latest step in controlling online communications
Studies: AI chatbots can influence voters
LG Elec says Microsoft and LG affiliates pursuing cooperation on data centres
Apple appoints Meta's Newstead as general counsel amid executive changes
AI's rise stirs excitement, sparks job worries
Australia's NEXTDC inks MoU with OpenAI to develop AI infrastructure in Sydney, shares jump
SentinelOne forecasts quarterly revenue below estimates, CFO to step down

Others Also Read