Search AI Products and News

Meta recently disclosed comprehensive technical details of the hardware for its Aria Gen2 research glasses, providing the most detailed technical analysis since its release in February this year. Compared to the initial product released in 2020, Aria Gen2 has achieved a comprehensive technological leap. The hardware design has been fully upgraded, with the new glasses weighing between 74-76 grams and offering eight different frame designs to accommodate various face shapes. The foldable temples enhance portability, while the 6-8 hour battery life meets daily usage requirements.

According to foreign media reports, Perplexity AI is in advanced negotiations for a funding round of up to $500 million, which is expected to push the company's valuation up to $14 billion. The round is being led by renowned venture capital firm Accel, marking the rapid rise of Perplexity in the AI search sector and its aim to challenge Google and Microsoft-backed OpenAI's dominance in search. AIbase observed that this news has sparked extensive discussions within the tech community, with investors showing extremely high enthusiasm for AI's potential to reshape the search market. Valuation surged

Giant Network announced that its social deduction game, Space Kill, has officially integrated the DeepSeek large language model and launched a native gameplay mode called "Imposter Challenge" based on this technology. This mode is currently undergoing gray testing and will be rolled out to all users later. This marks the industry's first deep application of DeepSeek's capabilities to innovate game mechanics, using AI to drive core gameplay and reshape the competitive and interactive experience.

A recent report by The Decoder on OpenAI's latest model, GPT-4.5, has sparked industry debate regarding its cost-effectiveness. While official data shows performance improvements over GPT-4, the cost has significantly increased. Specifically, GPT-4.5 outperforms GPT-4 in various aspects, including a 63.2% improvement in professional queries, a 57% improvement in everyday queries, and...

A recent study released by the University of Munich, the Munich Machine Learning Center, and Adobe Research shows that 12 top AI language models, including GPT-4o, Gemini1.5Pro, and Llama-3.3-70B, face significant performance degradation in long-text conceptual reasoning tasks. Despite these models supporting context processing of at least 128,000 tokens, their ability to establish deep logical connections still has fundamental limitations. The research team developed NOLIMA (No Text Matching).

OpenAI recently announced the detailed inference process of its latest inference model, o3-mini, which is seen as a response to the increasing pressure from competitors like DeepSeek-R1. This change marks a significant shift in OpenAI's strategy towards model transparency. Previously, OpenAI regarded the 'Chain of Thought' (CoT) as a core competitive advantage and chose to keep it concealed. However, with open models like DeepSeek-R1 fully showing their inference paths, this has led to a reevaluation.

Supported by OpenAI CEO Sam Altman, biotech startup Retro Biosciences is in the midst of a $1 billion Series A funding round, furthering its ambition to extend human lifespan. Previously, Altman invested $180 million in seed funding for the company. Retro Biosciences aims not only to extend human lifespan but also to prolong healthy lifespan, ensuring people remain in good health towards the end of their lives.

The research team from The Chinese University of Hong Kong (Shenzhen) and Shenzhen Big Data Institute has recently launched a medical large language model (LLM) named HuatuoGPT-o1. This model is specifically designed for complex reasoning in the medical field, aiming to enhance the reliability of medical diagnosis and decision-making. Unlike previous LLMs that focused on mathematical reasoning, HuatuoGPT-o1 specializes in the medical domain by simulating the rigorous thought processes of doctors in real-world practice, paving new avenues for the development of medical AI.