DeepSeek Updates Logical Reasoning Model R1 Amid Rising AI Rivalry - Jordan News
DeepSeek Updates Logical Reasoning Model R1 Amid Rising AI Rivalry Chinese AI startup DeepSeek released an update to its logical reasoning model R1 on Thursday morning, intensifying competition with U.S. counterparts like OpenAI. اضافة اعلان The updated model, titled R1-0528, was made available on the developer platform Hugging Face, although no official announcement or documentation has yet been published by DeepSeek. The company has not released benchmark comparisons or detailed descriptions of the model's capabilities. However, according to the LiveCodeBench leaderboard—a benchmark developed by researchers at UC Berkeley, MIT, and Cornell—the updated R1 model ranks just below OpenAI's O4-mini and O3 models in code generation tasks, while outperforming Grok-3-mini from XAI and Qwen 3 from Alibaba. According to Bloomberg, a DeepSeek representative described the release on a WeChat group as a "simple experimental update," encouraging users to begin testing the model. Meanwhile, Alibaba also announced an updated version of its flagship Qwen AI model, reflecting the accelerating pace of innovation in China's AI sector. DeepSeek made headlines earlier this year with the launch of the original R1, challenging assumptions that U.S. export restrictions would stifle China's AI progress. The model received widespread praise for achieving performance on par with top Western models—at significantly lower computational cost. The release triggered a global tech selloff and renewed debate around whether cutting-edge AI requires massive capital and compute resources. In response to DeepSeek's rapid ascent, major Chinese tech firms like Alibaba and Tencent launched rival models claiming superiority. Meanwhile, OpenAI and Google revised their pricing strategies, with OpenAI introducing the lightweight o3 mini model to reduce compute requirements. DeepSeek, headquartered in Shanghai, is widely expected to release its next-generation R2 model soon. According to sources cited by Reuters in March, the R2 release is scheduled for May. The company also introduced an enhanced multilingual version of its V3 model earlier this year.
— Agencies
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Jordan News
17 minutes ago
- Jordan News
Google Launches App to Run AI Models Without Internet - Jordan News
Last week, Google quietly released a new app that allows users to run a variety of artificial intelligence models—available through the Hugging Face platform—directly on their smartphones, without needing an internet connection. اضافة اعلان The app, named Google AI Edge Gallery, is currently available for Android devices and is expected to launch soon on iOS. It enables users to search, download, and run compatible AI models offline by leveraging supported phone processors. While cloud-based AI models typically offer more power, they come with certain drawbacks. Some users may be reluctant to send personal or sensitive data to remote data centers, or may prefer to access AI features without relying on Wi-Fi or mobile networks. Google has labeled the app as an "Alpha release," and it can be downloaded from GitHub by following the published setup instructions. The app's home screen features a shortcut to Google's Gemma 3n model. A notable feature in the app is the 'Prompt Lab', which allows users to perform simple tasks like summarizing or rephrasing text using templates and customizable settings to fine-tune model behavior. Google warns that performance may vary across devices: newer smartphones with more powerful components will run models faster. However, model size also plays a crucial role—larger models take more time to complete tasks like answering questions about an image compared to smaller ones. Google is inviting the developer community to share feedback on their experience with the Google AI Edge Gallery. The app is released under the Apache 2.0 license, allowing for wide use, including in commercial contexts, without significant restrictions.
Jordan News
3 days ago
- Jordan News
DeepSeek Updates Logical Reasoning Model R1 Amid Rising AI Rivalry - Jordan News
DeepSeek Updates Logical Reasoning Model R1 Amid Rising AI Rivalry Chinese AI startup DeepSeek released an update to its logical reasoning model R1 on Thursday morning, intensifying competition with U.S. counterparts like OpenAI. اضافة اعلان The updated model, titled R1-0528, was made available on the developer platform Hugging Face, although no official announcement or documentation has yet been published by DeepSeek. The company has not released benchmark comparisons or detailed descriptions of the model's capabilities. However, according to the LiveCodeBench leaderboard—a benchmark developed by researchers at UC Berkeley, MIT, and Cornell—the updated R1 model ranks just below OpenAI's O4-mini and O3 models in code generation tasks, while outperforming Grok-3-mini from XAI and Qwen 3 from Alibaba. According to Bloomberg, a DeepSeek representative described the release on a WeChat group as a "simple experimental update," encouraging users to begin testing the model. Meanwhile, Alibaba also announced an updated version of its flagship Qwen AI model, reflecting the accelerating pace of innovation in China's AI sector. DeepSeek made headlines earlier this year with the launch of the original R1, challenging assumptions that U.S. export restrictions would stifle China's AI progress. The model received widespread praise for achieving performance on par with top Western models—at significantly lower computational cost. The release triggered a global tech selloff and renewed debate around whether cutting-edge AI requires massive capital and compute resources. In response to DeepSeek's rapid ascent, major Chinese tech firms like Alibaba and Tencent launched rival models claiming superiority. Meanwhile, OpenAI and Google revised their pricing strategies, with OpenAI introducing the lightweight o3 mini model to reduce compute requirements. DeepSeek, headquartered in Shanghai, is widely expected to release its next-generation R2 model soon. According to sources cited by Reuters in March, the R2 release is scheduled for May. The company also introduced an enhanced multilingual version of its V3 model earlier this year. — Agencies
Ammon
3 days ago
- Ammon
China's DeepSeek releases an update to its R1 reasoning model
Ammon News - Chinese artificial intelligence startup DeepSeek released an update to its R1 reasoning model in the early hours of Thursday, stepping up competition with U.S. rivals such as OpenAI. DeepSeek launched R1-0528 on developer platform Hugging Face, but has yet to make an official public announcement. It did not publish a description of the model or comparisons. But the LiveCodeBench leaderboard, a benchmark developed by researchers from UC Berkeley, MIT, and Cornell, ranked DeepSeek's updated R1 reasoning model just slightly behind OpenAI's o4 mini and o3 reasoning models on code generation and ahead of xAI's Grok 3 mini and Alibaba's Qwen 3.
