Search AI Products and News

Recently, Tencent Cloud officially launched the API service for the Tencent Hunyuan A13B model on its official website. The input price is set at 0.5 yuan per million Tokens, and the output price is 2 yuan per million Tokens, which has quickly sparked enthusiastic discussions in the developer community. As the first 13B-level MoE (Mixture of Experts) open-source hybrid inference model in the industry, Hunyuan-A13B features a total of 80B parameters and only 13B activated parameters, achieving performance comparable to leading open-source models of the same architecture, while also demonstrating efficient reasoning capabilities.

In August 2024, iFlytek officially launched the Starfire Ultra Human-like Interactive Technology. Through end-to-end speech modeling and multi-dimensional emotional disentanglement training, it achieves three core breakthroughs: response speed, emotional resonance, and controllable speech expression. This technology can accurately detect emotional fluctuations in user speech and respond with appropriate tone in real time, while supporting dynamic adjustment of speech rate, voice, and character settings. It marks a significant leap from 'functional implementation' to 'emotional connection' in voice interaction. Currently, the Super Human-like Interactive API has been officially launched on the iFlytek Open Platform, allowing developers to access the technology at a low cost.

Welcome to the AIbase [AI Daily Report] section! Spend three minutes a day to learn about the latest AI events, helping you understand AI industry trends and innovative AI product applications. For more AI news, visit: https://www.aibase.com/zh1. Baidu officially releases the WENXIN Large Model 4.5 series and fully opens it to the public, featuring ten new models with various parameter configurations. These models are trained and inferred using the PaddlePaddle framework, achieving a FLOPs utilization rate of 47%, and perform well in multi-modal text tasks.

Recently, Google announced that the API of its flagship AI model, Gemini2.5Pro, has been reintroduced to the free tier of Google AI Studio. This news has triggered widespread attention and enthusiastic discussions within the developer community. According to AIbase, this move marks another important advancement in Google's efforts to popularize AI technology, offering developers lower barriers to innovation. As the most advanced AI model from Google so far, Gemini2.5Pro is known for its exceptional multimodal capabilities and strong reasoning power.

Recently, the Qwen VLo multimodal large model was officially released, achieving significant advancements in image content understanding and generation, offering users a brand-new visual creation experience. According to the introduction, Qwen VLo has been comprehensively upgraded based on the advantages of the original Qwen-VL series models. The model not only can accurately understand the "world", but also can perform high-quality re-creation based on understanding, truly achieving the transition from perception to generation. Users can now access Qwen Chat (chat.qwen.ai)

OpenAI announced the official opening of API access to its deep research models, providing developers with a powerful set of tools including automatic web search, data analysis, MCP (Model Communication Protocol), and code execution. The models opened include the deep research versions of o3 and o4-mini, which have already been used in ChatGPT. Now, developers can directly call these models via API. These models are particularly suitable for complex tasks that require obtaining the latest information and performing advanced reasoning. In terms of function expansion, o3,

Recently, Google officially launched its latest text-to-image model **Imagen4** through the Gemini API, marking an important milestone in the field of generative AI (AIGC). According to Google's official blog and community feedback, Imagen4 has achieved breakthroughs in generating text within images, solving a long-standing technical bottleneck in AIGC, and providing developers with a tool for creating high-quality visual content. It is reported that the model comes in two versions: **Imagen4** and **Imagen4Ultra**, with respective pricing details yet to be fully disclosed.

On June 19th, the Tongyi APP has comprehensively upgraded its translation capabilities, covering four core scenarios: text translation, simultaneous interpretation translation, document translation, and image translation, creating the strongest translation complex for individual and professional users. After the upgrade, the translation capabilities support 119 languages and dialects, achieving comprehensive improvements in accuracy, professionalism, and interaction experience. Whether it's cross-border office work, academic reading, or travel, the Tongyi APP can provide a truly all-scenario and all-modal translation solution. The Tongyi APP now supports 119 languages and dialects.

Recently, AI Infrastructure company SiliconBase Flow officially announced a strategic cooperation with Alibaba Cloud. SiliconBase Flow has joined the "Alibaba Cloud Flower Program" and entered the Alibaba Cloud Marketplace. This collaboration marks the beginning of deep cooperation in the AI field, working together to promote the development of large model services. According to the cooperation content, the core API services of SiliconBase Flow's large model inference platform, SiliconCloud, will fully integrate into Alibaba Cloud Bailian Platform, supported by Alibaba Cloud Lingjun Intelligent Computing Clusters, providing convenient and efficient large model services for cloud customers.

The plan's advancement has triggered dissatisfaction among developers. According to VentureBeat, OpenAI's current API deprecation involves the GPT-4.5 Preview version, which is an intermediary model launched between GPT-4 and GPT-5. Multiple developers have stated that this sudden change has disrupted their development plans, forcing significant adjustments to applications already built based on this API. A developer who wished to remain anonymous commented: We understand the necessity of technological iteration, but the lack of sufficient buffer time in the schedule has brought additional pressure to development teams.

With the surge in demand for high-quality data from artificial intelligence, efficient web scraping tools have become the core need for AI developers. AIbase learned from social media platforms that Firecrawl has launched a groundbreaking feature - the new /search API, which allows both web search and content scraping with just one API call and outputs data in AI-friendly formats. This release marks an important step forward for Firecrawl in the field of AI-driven web data processing. This article will explore

Recently, the Natural Language Intelligence Team of the Tongyi Lab officially released and open-sourced VRAG-RL — a multi-modal RAG reasoning framework driven by visual perception. It aims to solve the problem of how AI can retrieve key information and conduct fine-grained reasoning from visual languages such as images, tables, design drafts, etc., in real-world business scenarios. Retrieving and reasoning about key information in complex visual document knowledge bases has always been a major challenge in the AI field. Traditional retrieval-augmented generation (RAG) methods struggle when handling visually rich information because they find it difficult to...

In May, Quark officially launched the new "Advanced Research" feature and began inviting users for limited early access. This function is based on the Tongyi Qianwen large model, supporting complex topics such as academic research subjects and industry analysis, completing the entire process from material collection, data analysis, viewpoint extraction to report generation. It enables the user to input a topic and output a finished product. Users can now apply for an invitation code through the Quark App or PC version, activate their eligibility, and then click the 'Advanced Research' icon on the home page to submit their requests. The system will automatically generate a structured report within a few minutes and supports exporting it as a PDF.