Latest news with #DeepSeekV3
Miami Herald
4 days ago
- Business
- Miami Herald
Atlas Cloud Launches High-Efficiency AI Inference Platform, Outperforming DeepSeek
Developed with SGLang, Atlas Inference surpasses leading AI companies in throughput and cost, running DeepSeek V3 & R1 faster than DeepSeek themselves. NEW YORK CITY, NEW YORK / ACCESS Newswire / May 28, 2025 / Atlas Cloud, the all-in-one AI competency center for training and deploying AI models, today announced the launch of Atlas Inference, an AI inference platform that dramatically reduces GPU and server requirements, enabling faster, more cost-effective deployment of large language models (LLMs). Atlas Inference, co-developed with SGLang, an AI inference engine, maximizes GPU efficiency by processing more tokens faster and with less hardware. When comparing DeepSeek's published performance results, Atlas Inference's 12-node H100 cluster outperformed DeepSeek's reference implementation of their DeepSeek-V3 model while using two-thirds of the servers. Atlas' platform reduces infrastructure requirements and operational costs while addressing hardware costs, which represent up to 80% of AI operational expenses. "We built Atlas Inference to fundamentally break down the economics of AI deployment," said Jerry Tang, Atlas CEO. "Our platform's ability to process 54,500 input tokens and 22,500 output tokens per second per node means businesses can finally make high-volume LLM services profitable instead of merely break-even. I believe this will have a significant ripple effect throughout the industry. Simply put, we're surpassing industry standards set by hyperscalers by delivering superior throughput with fewer resources." Atlas Inference's performance also exceeds major players like Amazon, NVIDIA and Microsoft, delivering up to 2.1 times greater throughput using 12 nodes compared to competitors' larger setups. It maintains sub-5-second first-token latency and 100-millisecond inter-token latency with more than 10,000 concurrent sessions, ensuring a scaled, superior experience. The platform's performance is driven by four key innovations: Prefill/Decode Disaggregation: Separates compute-intensive operations from memory-bound processes to optimize efficiencyDeepExpert (DeepEP) Parallelism with Load Balancers: Ensures over 90% GPU utilizationTwo-Batch OverlapTechnology: Increases throughput by enabling larger batches and utilization of both compute and communication phases simultaneouslyDisposableTensor Memory Models: Prevents crashes during long sequences for reliable operation "This platform represents a significant leap forward for AI inference," said Yineng Zhang, Core Developer at SGLang. "What we built here may become the new standard for GPU utilization and latency management. We believe this will unlock capabilities previously out of reach for the majority of the industry regarding throughput and efficiency." Combined with a lower cost per token, linear scaling behavior, and reduced emissions compared to leading vendors, Atlas Inference provides a cost-efficient and scalable AI deployment. Atlas Inference works with standard hardware and supports custom models, giving customers complete flexibility. Teams can upload fine-tuned models and keep them isolated on dedicated GPUs, making the platform ideal for organizations requiring brand-specific voice or domain expertise. The platform is available immediately for enterprise customers and early-stage startups. About Atlas Cloud Atlas Cloud is your all-in-one AI competency center, powering leading AI teams with safe, simple, and scalable infrastructure for training and deploying models. Atlas Cloud also offers an on-demand GPU platform that delivers fast, serverless compute. Backed by Dell, HPE, and Supermicro, Atlas delivers near instant access to up to 5,000 GPUs across a global SuperCloud fabric with 99% uptime and baked-in compliance. Learn more at SOURCE: Atlas Cloud press release
Indian Express
08-05-2025
- Business
- Indian Express
Mistral announces new AI model Medium 3 at 8x lower cost
French AI startup Mistral has introduced a frontier-level AI model, Mistral Medium 3. The new model from the Paris-based AI company is said to have outperformed models like Claude Sonnet 3.7 and GPT-4o on numerous benchmarks. The new model reportedly costs less than DeepSeek V3. The company has said that organisations can use the new model through its new AI assistant called Le Chat Enterprise that features an agent builder and allows full integration with a variety of apps. Mistral has also teased a more powerful model which will be introduced in the coming weeks. Mistral Medium 3 is said to be pushing efficiency and usability of language models even further. Mistral claims that the new Medium 3 brings a new class of models that balances state-of-the-art performance, is 8x lower in cost, and offers simple deployability to accelerate enterprise usage. The model also leads in professional use cases like coding and multimodal understanding. When it comes to enterprise capabilities, Medium 3 offers hybrid or on-premises in-VPC deployment, custom post-training, and allows integration into enterprise tools and systems. According to the company, the model performs at or above 90 per cent of Claude Sonnet 3.7 on benchmarks across the board at a considerably lower cost – $0.4 input/$2 output per M token. Medium 3 has also surpassed models such as Llama 4 Maverick and enterprise models like Cohere Command A. When it comes to pricing in terms of API and self-deployed systems, the model beats DeepSeek V3. It can also be deployed on any cloud, including self-hosted environments of four GPUs and above. The company claims the model is designed to be frontier-class, particularly in categories of professional use. When it comes to benchmarks, Mistral Medium 3 delivers top performance in instruction following (ArenaHard: 97.1%) and math (Math500: 91%), with strong results in long context tasks (RULER 32K: 96%). In terms of human evaluations, Medium 3 outperforms competitors, especially in coding. The model beats Claude Sonnet 3.7, DeepSeek 3.1, and GPT-4o in several cases.
South China Morning Post
25-03-2025
- Business
- South China Morning Post
DeepSeek's upgraded foundational model excels in coding and maths
Chinese artificial intelligence (AI) star DeepSeek has upgraded its open-source V3 large language model by adding parameters and improving capabilities in coding and solving mathematical problems. Advertisement The DeepSeek-V3-0324, named after its predecessor and the launch date, has 'enhanced reasoning capabilities, optimised front-end web development and upgraded Chinese writing proficiency', according to a notice on the company's website. The new version and DeepSeek V3 are both foundation models trained on vast data sets that can be applied in different use cases, including that of a chatbot. DeepSeek R1, the reasoning model, is based on DeepSeek V3. The updated foundation model has made improvements in several benchmarks, especially the American Invitational Mathematics Examination (AIME), where it scored 59.4 compared with 39.6 for its predecessor, while achieving an increase of 10 points on LiveCodeBench to achieve 49.2, DeepSeek data showed. This illustration photograph taken on January 29, 2025 shows screens displaying the logos of DeepSeek and OpenAI's AI chatbot ChatGPT. Photo: AFP Compared with DeepSeek V3, which has 671 billion parameters and adopts the company's own commercial license, the new 685-billion-parameter model uses the MIT software licence that is the most popular on developer platform GitHub. Advertisement Launched on AI community Hugging Face as well as the company's own website, DeepSeek-V3-0324 is now the top trending model on Hugging Face, receiving positive comments on its performance.
Yahoo
14-02-2025
- Business
- Yahoo
China's Xi Jinping to hold high-level meeting on Monday with top tech entrepreneurs
Chinese President Xi Jinping will host a meeting next week with some of the nation's top entrepreneurs - including six Hangzhou-based start-ups known as the "Six Little Dragons" - to recognise progress in critical areas of technological advancement and show support to the private sector, according to sources. Between 20 and 30 founders and chief executives from China's largest technology companies were expected to assemble in Beijing on Monday, according to the sources, who spoke on condition of anonymity. The meeting comes at a critical time for China before the national legislature and the political advisory body enter their annual meetings to formulate the nation's strategies to regain its economic growth pace and chart a course through the evolving trade war and technological race with the US. The "two sessions" process is also expected to legislate to better protect the private sector, the Ministry of Justice said last October. Do you have questions about the biggest topics and trends from around the world? Get the answers with SCMP Knowledge, our new platform of curated content with explainers, FAQs, analyses and infographics brought to you by our award-winning team. Against the headwinds, companies such as DeepSeek and Huawei have broken through with new services and products that have mitigated US restrictions, the sources said. Chinese President Xi Jinping (left, front) in a meeting during a symposium on private enterprises at the Great Hall of the People in Beijing on November 1, 2018. Photo: Xinhua alt=Chinese President Xi Jinping (left, front) in a meeting during a symposium on private enterprises at the Great Hall of the People in Beijing on November 1, 2018. Photo: Xinhua> DeepSeek in December launched a low-cost large language model (LLM) with 671 billion parameters called the DeepSeek V3, which it claimed was trained in around two months for US$5.58 million. The feat by the Hangzhou-based company seemed to challenge Silicon Valley's dominance in artificial intelligence, which thus far has been led by US companies such as OpenAI, Microsoft and Google. China's "DeepSeek moment" and its implications will be "good for innovation because making AI less costly and more accessible means more companies and developers can participate in the upside of AI development, and more consumers will benefit from the proliferation of useful and cool applications", Alibaba Group Holding's co-founder and chairman Joe Tsai recently wrote in an op-ed column in the Post. Alibaba owns the Post. DeepSeek's breakthrough was matched - surpassed, in certain parameters - by other Chinese companies. Alibaba's latest open-source Qwen artificial intelligence model called the 2.5-Max beat DeepSeek V3 to become the top-ranked non-reasoning model from a Chinese developer, according to a third-party benchmarking and ranking platform. Alibaba this week struck an agreement to supply Qwen to run on Apple's iPhones in China, an endorsement of the company's AI prowess. Advances have been made elsewhere to break out of America's technological chokehold. Huawei surprised the world in September 2023 when it launched its Mate 60 Pro flagship smartphone with its own Kirin 9000s processor that was capable of supporting 5G connectivity. The move broke through the 2020 US export restrictions that prevented Huawei from obtaining advanced integrated circuits from major contract chipmakers around the world. Unitree's H1 humanoid robots appeared in the 2025 Lunar New Year Gala performance on CGTN on Jan 28, 2025. Photo: Unitree alt=Unitree's H1 humanoid robots appeared in the 2025 Lunar New Year Gala performance on CGTN on Jan 28, 2025. Photo: Unitree> The invited companies are understood to include Huawei Technologies, Tencent Holding, Xiaomi, DeepSeek, Alibaba and Unitree Robotics, the sources said. Spokespeople from Huawei, Tencent, Xiaomi, DeepSeek, Alibaba and Unitree declined to comment. Reuters first reported the Monday meeting, citing unidentified sources. Among other invited companies are the so-called Six Little Dragons of China's technology prowess: half a dozen start-ups based in the Zhejiang provincial capital of Hangzhou. Besides DeepSeek, they are the robot maker Unitree, Deep Robotics, the video game studio Game Science, the brain-machine interface innovator BrainCo and the 3D interior design software developer Manycore. This article originally appeared in the South China Morning Post (SCMP), the most authoritative voice reporting on China and Asia for more than a century. For more SCMP stories, please explore the SCMP app or visit the SCMP's Facebook and Twitter pages. Copyright © 2025 South China Morning Post Publishers Ltd. All rights reserved. Copyright (c) 2025. South China Morning Post Publishers Ltd. All rights reserved.
South China Morning Post
14-02-2025
- Business
- South China Morning Post
President Xi to meet China's tech entrepreneurs in nod to breakthroughs
Published: 5:45pm, 14 Feb 2025 Chinese President Xi Jinping will host a meeting next week with some of the nation's top entrepreneurs to recognise progress in critical areas of technological advancement and show support to the private sector, according to sources. Between 20 and 30 founders and chief executives from China's largest technologies were expected to assemble in Beijing on Monday, according to the sources, who spoke on condition of anonymity. The meeting comes at a critical time for China before the legislature and the advisory body enter their annual meetings to formulate the nation's strategies to regain its economic growth pace and chart a course through the evolving trade war and technological race with the US. The ' two sessions ' may also legislate to better protect the private sector, the Ministry of Justice said last October . Against the headwinds, companies such as DeepSeek and Huawei have broken through with new services and products that have mitigated US restrictions, the sources said. Chinese President Xi Jinping (L, front) in a meeting during a symposium on private enterprises at the Great Hall of the People in Beijing on November 1, 2018. Photo: Xinhua. DeepSeek in December launched a low-cost large language model (LLM) with 671 billion parameters called the DeepSeek V3, which it claimed was trained in around two months for US$5.58 million. The feat by the Hangzhou-based company seemed to challenge Silicon Valley's dominance in artificial intelligence , which thus far has been led by US companies such as OpenAI, Microsoft and Google.
