Search AI Products and News

Waymo announced that it will launch an autonomous taxi service in Dallas next year, marking another key city deployment following Los Angeles and San Francisco. The company has formed a strategic partnership with Avis Budget Group, which will handle the charging and maintenance of the all-electric Jaguar I-Pace fleet. Waymo has completed preliminary testing in Dallas and plans to gradually expand its fleet to hundreds of vehicles. Currently, Waymo is operating commercially in five U.S. cities and plans to expand to Washington D.C. and Miami next year. This collaboration

The SWE-bench team has launched a lightweight open-source project called mini-SWE-agent, which can fix 65% of GitHub project bugs with just 100 lines of code. This version simplifies the architecture, removes complex tool interfaces, and executes commands through Bash, offering better compatibility and fewer dependencies. While retaining the core functions of the original version, it provides tools such as batch inference and a trajectory browser, supporting command-line startup and visual monitoring. The team states that mini-SWE-agent is suitable for developers who prefer simplicity, while the original version is more appropriate for others.

Ali International AI Solution Marco wins the SAIL Star Award at the 2025 World Artificial Intelligence Conference. The solution supports more than 40 languages and covers over 60 e-commerce scenarios, with a hallucination rate of only 0.02%, significantly better than industry standards. Ali International AI service has a daily call volume exceeding 1 billion, and has been opened to external ecosystems, driving a 23-fold increase in AI call volume for partners. It has now launched three Agent applications: Smart Refund, Customs Code, and Online Recruitment, achieving a 15% reduction in refund costs, a 23% improvement in accuracy, and improved recruitment efficiency.

iFlytek's upgraded Spark X1 model enhances translation, reasoning, and text generation, achieving top-tier global performance with 20% better translation quality. It scores over 80 in EN-CN conference translation, reduces AI hallucinations, and boosts efficiency for 100+ firms by 50%. Available via web, app, and API.....

The Qwen-MT machine translation model is officially released. Qwen-MT is based on the powerful Qwen 3 model, and has been trained on a massive amount of multilingual and translation data. It combines reinforcement learning technology to significantly improve the accuracy and fluency of translation results. Developers can experience its fast and accurate translation capabilities directly through the Qwen API (qwen-mt-turbo). The core highlights of Qwen-MT include supporting bidirectional translation between 92 languages, covering more than 95% of the global population, and meeting a wide range of language communication needs.

The ByteDance Seed team officially announced its latest research achievement - the Seed LiveInterpret 2.0 end-to-end simultaneous translation large model. The release of this model marks a major breakthrough in machine simultaneous translation technology, with its translation accuracy approaching that of professional interpreters, and an extremely low latency of only 3 seconds. It also features real-time voice cloning, allowing the translated speech to be output in the speaker's tone, greatly enhancing the naturalness and fluency of cross-language communication.

Metasearch offers a cost-effective Bing API alternative with multilingual index, multi-modal search, and developer-friendly APIs at $0.03/query.....

OpenAI and Oracle have reached a $30 billion data center cooperation agreement, involving 4.5 gigawatts of power capacity (equivalent to the output of two Hoover Dams). This deal is part of their Star Gate project, with construction taking place in Abilene, Texas. Oracle has spent nearly $5 billion on data centers in the past two years, supporting not only OpenAI but also existing customers. Notably, OpenAI's annual revenue of $10 billion is only one-third of this contract's value. This collaboration has driven Oracle's stock to a new high, and its founder Ellison briefly became the world's richest person.

ByteDance open sources the lightweight multilingual translation model Seed-X, which supports bidirectional translation in 28 languages and demonstrates performance comparable to top-level large models. This 7-billion-parameter model is based on the Mistral architecture and focuses on translation optimization, showing excellent performance in multiple areas. It uses an innovative training strategy to generate high-quality data and optimize deployment efficiency. This is another open-source project from ByteDance following BAGEL and Seed-Coder, promoting advancements in AI translation technology.

OpenAI announced a 100x expansion plan, aiming to deploy over 1 million GPUs by 2025. The core elements include: 1) The Stargate project, which involves an investment of $50 billion to build the world's largest AI training cluster in Texas; 2) Restructuring the chip supply chain; 3) Addressing energy challenges. The project is chaired by SoftBank CEO Masayoshi Son and involves collaborations with industry giants such as Oracle, Arm, and Microsoft. This strategic move will reshape the AI infrastructure landscape and drive the industry into a new phase of rapid development, expected to have a significant impact on the market.

Stability-AI's SPAR3D model revolutionizes 3D reconstruction from single images in 0.7s, combining regression and generative modeling via a two-stage point sampling and meshing architecture. It outperforms traditional methods in geometric accuracy and texture quality on GSO and OmniObject3D datasets. Open-sourced.....

MirageLSD, the world's first AI real-time video conversion model with 40ms latency, enables instant scene/outfit changes via gestures. Applied in gaming/livestreaming, its LSD model uses Diffusion Forcing to eliminate long-generation errors.....

Baidu AI Assistant introduces 'Video Call' feature, enabling real-time video chats via Baidu APP. It covers four scenarios: object/insect identification, styling advice, pet behavior analysis, and emotional support, with dialect recognition for elderly users. This upgrade transforms AI from a tool to a smart life companion.....

Canadian ATV riders lost in Mabou Forest for 5 hours used ChatGPT for real-time navigation with weak signals. The AI analyzed GPS coordinates and terrain to guide them safely, demonstrating AI's potential in outdoor rescue.....

The neighborhood social application Nextdoor has launched a major update with three new features: local news aggregation (collaborating with 3,500 local media outlets), real-time safety alerts (weather, traffic, and disaster warnings), and AI-powered Faves recommendations (personalized local service suggestions generated based on 15 years of neighborhood data). The platform aims to address previous issues with incorrect information by improving content quality and timeliness to revitalize user engagement. CEO Tolia emphasized that its core advantage lies in digital neighborhood reputation, which is a unique value that other platforms cannot replicate.

TRAE.ai launches Kimi-K2 model and Grok-4(Beta). Kimi-K2 excels in code/math with MoE architecture, rivaling GPT-4.1. Easy 3-step access. International version adds Grok-4(Beta) testing alongside Claude, Gemini, GPT.....

AI's 'time travel' trend thrives as ChatGPT transforms childhood photos. TikTok's AI effect drew 170K users, but results vary: Musk's image was unrecognizable, Asian stars distorted, while Eddie Peng fared slightly better. Experts note AI predicts general trends, not individual changes, yet this playful tech sparks social media buzz.....

Meta and Berkeley developed StreamDiT for real-time AI video generation: 1) 16fps 512p on single GPU, 4B-param model creates 1-min videos with live edits; 2) Novel buffer enables parallel processing (2 frames/0.5s) with 8-step optimization; 3) Trained on 3K HD videos + 2.6M dataset; 4) Outperforms rivals in motion smoothness, 30B-param version shows quality potential; 5) Enables interactive video despite transition flaws.....

Google DeepMind open sources the GenAI Processors Python library, helping developers build efficient generative AI workflows. The library supports asynchronous processing of multimodal data and optimizes Gemini API application development, significantly reducing latency in real-time applications. Core features include a modular Processor interface, streaming API design, and concurrency optimization, enabling rapid development of real-time applications such as intelligent assistants. Currently only supports Python, but with an open community contribution model, future plans include expanding functionality to cover more scenarios.