Alibaba offers DeepSeek on cloud service after Microsoft, Amazon and Huawei
Alibaba Cloud users can log into its PAI Model Gallery, where DeepSeek’s latest models and ‘distilled’ versions of these products are found
When you spot a great trading opportunity but don’t have the funds ready, use contra trading limits to buy stocks or ETFs immediately without upfront cash. Trade now and settle later with the Tiger Cash Boost Account. Enjoy up to 7 days interest-free with a $20K credit limit. Plus, trade commission-free for up to 6 months, perfect for short-term trading. Plus, sign up and deposit to receive rewards up to $1000 and enhance your investment journey!
Alibaba Group Holding’s cloud-computing services unit on Monday made available DeepSeek’s artificial intelligence (AI) models on its platform, following moves by other Big Tech companies to bring the Chinese start-up’s open-source systems to their clients.
“On [our] platform, users can achieve the whole process from training to deployment to inference with zero coding,” Alibaba Cloud said in a statement posted on WeChat. Alibaba owns the South China Morning Post.
It said the platform simplifies the model-development process, “bringing developers and enterprise users a faster, more efficient and more convenient AI development and application experience”.
Alibaba Cloud users can log into its PAI Model Gallery – a collection of open-source large language models (LLMs) – where they can select DeepSeek’s AI models and deploy them to power their own reasoning and text-generating applications, according to the statement.
The gallery contains the Hangzhou-based start-up’s most advanced AI models, DeepSeek-V3 and DeepSeek-R1, touted as having been developed at a fraction of the cost and computing power typically spent by major AI tech companies to build LLMs. It also provides so-called distilled versions of those models, such as DeepSeek-R1-Distill-Qwen-7B.
LLMs are the technology behind generative AI services like OpenAI’s ChatGPT. Open source gives public access to a software program’s source code, allowing third-party developers to modify or share its design, fix broken links or scale up its capabilities.
Distillation is a means of training smaller models to mimic the behaviour of larger, more sophisticated models, and at the same time significantly reduce computational costs. The practice is common at many companies looking to scale down the size of their models, while offering similar performance to users.
The latest move by Alibaba Cloud, which last month released its new Qwen 2.5-Max model that rivals DeepSeek-V3, reflects a growing trend among major tech companies to support the start-up’s models for the benefit of their own customers.
The cloud-computing unit of Huawei Technologies, for example, worked overtime during the Lunar New Year break with AI infrastructure start-up SiliconFlow to make DeepSeek’s V3 and R1 models available for users on the Shenzhen-based firm’s Ascend platform.
The platform’s performance matched how “DeepSeek models run on global premium graphic processing units”, Huawei Cloud said in a statement on Saturday.
Cloud computing technology enables enterprises to manage or distribute over the internet a range of software and other digital resources as an on-demand service, just like electricity from a power grid. These resources are stored inside data centres.
Chinese social media and video gaming powerhouse Tencent Holdings has also started supporting DeepSeek’s R1 reasoning model on its cloud-computing platform, where users are able to execute an easy three-minute set-up process, the company said on Sunday.
Top AI chip designer Nvidia made DeepSeek-R1 available to users of its NIM microservice since Thursday, saying the model provides “state-of-the-art reasoning capabilities”, “high inference efficiency”, as well as “leading accuracy” for tasks requiring logical inference, reasoning, maths, coding and language understanding.
OpenAI investor Microsoft earlier last week launched R1 support on its Azure cloud-computing platform and developer platform GitHub, allowing clients to build AI applications that run locally on Copilot+ personal computers. E-commerce giant Amazon.com has enabled developers to create applications with R1 through Amazon Web Services.
Meanwhile, some experts question whether the significance of DeepSeek’s cost-effective AI model breakthrough might have been exaggerated.
Fudan University computer science professor Zheng Xiaoqing said the training expenditure for DeepSeek’s V3 model, which was in the start-up’s technical report, “excluded the costs associated with prior research and experiments on architectures, algorithms, or data”.
DeepSeek’s success stemmed from “engineering optimisation”, which “will not have a huge impact on chip purchases or shipments”, Zheng was quoted as saying in an interview with Chinese newspaper National Business Daily.
Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.
