Welcome to the AI Daily section! This is your daily guide to exploring the world of artificial intelligence. Every day, we bring you the hottest topics in the AI field, focusing on developers, helping you understand technology trends and discover innovative AI product applications.
Discover fresh AI products click here: https://top.aibase.com/
1. OpenAI Introduces Advanced Voice Mode for ChatGPT Plus Users
OpenAI has announced the launch of an advanced voice mode, offering a more natural and real-time conversational experience, allowing users to interrupt and perceive emotional responses. Test users will receive an email notification, with plans to make it accessible to all Plus users in the fall. Video and screen sharing features will also be introduced. GPT-4o has undergone multilingual testing, prioritizes privacy, and plans to release an evaluation report.
AiBase Highlights:
🔊 Advanced voice mode offers a more natural, real-time conversational experience, allowing users to interrupt and perceive emotional responses.
📧 Test users will receive instructions via email, with plans to fully open to Plus users in the fall.
🔗 OpenAI plans to release an evaluation report on the capabilities, limitations, and safety of GPT-4o.
2. OpenAI Releases New Model gpt-4o-64k-output-alpha, GPT-4o API Supports Up to 64K Output
OpenAI has introduced a new experimental model, gpt-4o-64k-output-alpha, with the highlight being the ability to output 64K tokens at once, providing richer and more detailed content, but at a higher price. This model meets the needs of users for longer text, supporting writing, programming, and data analysis. The cost for generating long texts is $18 per million output tokens, aimed at matching computational costs and encouraging rational use of the tool.
AiBase Highlights:
📈 GPT-4o model supports up to 64K output, suitable for users who need detailed content.
💰 Long text generation is expensive, costing $18 per million output tokens.
📝 The model opens up new possibilities in creative and research fields, promoting deeper communication and creation.
3. Midjourney V6.1 Released, Enhancing Distant Face Details
Midjourney has officially released version V6.1, bringing smarter, faster, clearer, and more detailed updates. This update breaks through image coherence, comprehensively improves image quality, accurately handles microscopic details, optimizes performance and speed, enhances text accuracy, upgrades personalized models, and introduces the --q2 mode, overall improving aesthetics.
AiBase Highlights:
✨ Breakthrough in image coherence: Handles complex structures more naturally and smoothly, boosting creative confidence.
🌟 Comprehensive improvement in image quality: Reduces pixel artifacts, enhances texture expressiveness, offering a broader range of stylistic expression.
🔍 Precise control over microscopic details: Handles small image features remarkably, enhancing the authenticity and expressiveness of the work.
Details link: https://top.aibase.com/tool/midjourneywangyeban
4. Xiaomi's "Big Model Xiao Ai" Fully Upgraded: Supports AI Image Editing and More
Xiaomi Corporation announces a major upgrade to the AI model version of Xiao Ai, covering smartphone, tablet, smart TV, smart speaker, and automotive product lines. The upgrade introduces innovative features such as AI image editing and external wake-up defense. The functional upgrades across product lines will enhance user experience and practicality.
AiBase Highlights:
🚀 Xiao Ai undergoes a comprehensive upgrade, covering multiple product lines, bringing innovative features.
💡 Smartphone focus on AI image editing, document Q&A, and smart video production.
📺 Smart TV offers movie knowledge Q&A, fitness guidance, and family planning.
5. Baidu Releases PaddleX3.0-beta Ascend Edition
PaddleX3.0-beta Ascend Edition is a cloud-edge collaborative low-code development tool launched by PaddlePaddle, aimed at helping developers solve practical problems in the industry at low cost and zero threshold. This version focuses on seven mainstream AI scenarios, offering rich model choices and low-code development methods, deeply adapted to Huawei's Ascend 910 chip, providing users with a convenient development paradigm.
AiBase Highlights:
🚀 Rich model selection, covering multiple task scenarios including image classification, object detection, text image layout analysis, etc.
⚙️ Low-code development method, unified API interface for full-process development, supports user-defined model process linkage.
💻 Deep adaptation to Huawei Ascend 910 chip, meeting diverse user needs.
Details link: https://github.com/PaddlePaddle/PaddleX/tree/release/3.0-beta
6. AI Virtual Clothing Model CatVTON: Small Parameters, Supports High Resolution
CatVTON is a small AI virtual clothing model suitable for fashion enthusiasts, featuring a lightweight network and low memory requirements, supporting high-resolution operations. The latest released code and deployment process allow users to easily experience virtual makeup technology at home, while also supporting deployment on ComfyUI and Gradio applications, making operations convenient.
AiBase Highlights:
🐈 Lightweight network, total parameters 899.06M, training parameters 49.57M, inference memory requirement below 8G, suitable for personal computer operation.
💻 Supports quick deployment on ComfyUI, simple steps to experience virtual makeup technology.
👗 Can perform inference on VITON-HD and DressCode datasets, supports multiple precision options, adapts to different hardware conditions.
Details link: https://top.aibase.com/tool/catvton
7. AI Hair Styling Tool Stable-Hair: Achieves Hair Style Transfer
Stable-Hair is an innovative hair transfer technology that solves the problem of traditional methods being unable to handle complex hairstyles, bringing users a brand-new hair trial experience. After extensive experimental verification, Stable-Hair has achieved industry-leading levels in hair transfer effects, enhancing the user's hair trial experience.
AiBase Highlights:
🌟 Stable-Hair solves the problem of traditional methods being unable to handle complex hairstyles.
👨🔬 Adopts a two-stage processing flow, first removing hair and then performing hair transfer.
🏆 Stable-Hair performs excellently in experiments, reaching industry-leading levels.
Details link: https://github.com/Xiaojiu-z/Stable-Hair
8. Fans Become Pure Love Warriors! Using AI to Embrace Xiang Liu Explodes on Social Platforms
This article discusses the application of artificial intelligence technology in the field of virtual emotional expression, taking the character Xiang Liu from the popular drama "Eternal Love 2" as an example. Fans interact with the character through AI-generated videos, showcasing a novel way of emotional release. However, the article also warns of the risks of over-immersing in the virtual world and the need to maintain emotional balance in real life.
AiBase Highlights:
🤖 AI technology helps fans express emotions through virtual interactions, innovative video generation methods spark a frenzy.
💑 AI couple photos are popular on social platforms, fans fulfill the dream of intimate contact with idols.
🧠 AI technology profoundly influences the entertainment field, providing a novel way to handle emotional attachment to fictional characters.
9. Meta Shuts Down Celebrity Chatbot Project, Focuses on AI Creation Tool AI Studio
Meta recently announced the closure of its celebrity chatbot project, shifting focus to AI Studio, a new tool that allows creators to make their own chatbots. Despite previous significant investments, the celebrity chatbot project did not meet expectations, but Meta remains hopeful about AI chatbots. CEO Mark Zuckerberg emphasizes the importance of AI chatbots and continues to promote the Meta AI assistant across multiple platforms.
AiBase Highlights:
🔥 Meta closes the celebrity chatbot project, focusing on the AI creation tool AI Studio.
💰 The company paid high fees for celebrity portraits, but the project did not meet expectations.
🤖 CEO Mark Zuckerberg emphasizes the importance of AI chatbots, continuing to promote the Meta AI assistant on multiple platforms.
10. TikTok Blogger "Makona" Strikes Again! AI Continues "Fairy Tale Town" with Abstract Lyrics Receiving 730K Likes
Recently, a song on the TikTok platform, re-lyricized and sung by AI, "Fairy Tale Town," has caused a sensation, receiving over 730,000 likes and becoming a hot topic of online discussion. The AI version's lyrics are filled with abstract imagery, the performance is outstanding, and it has sparked deep thinking. AI creation has led people to think about abstract thinking and creative logic. It also reflects the online culture's pursuit of novelty content, and AI creation needs to be viewed rationally. The AI version of "Fairy Tale Town" showcases the possibilities of AI in the creative field, increasing public interest in AI art creation.
AiBase Highlights:
⚡ AI re-lyricized "Fairy Tale Town" causes a sensation on TikTok, receiving over 730,000 likes, showing the potential of AI in the music creation field.
🎭 The AI version's lyrics are filled with abstract imagery, the performance is outstanding, sparking deep thinking about AI creativity.
🌐 The contemporary online culture's pursuit of novelty and humor content, AI creation needs to be viewed rationally, exploring the essence and boundaries of AI creativity.
11. Zhou Hongyi Announces 360 Security Large Model Free
Zhou Hongyi announces that 360 Group will open the 360 Security Large Model for free, aiming to popularize large model technology, promote technological progress and productivity transformation in the security industry, making it more affordable and effective for more enterprises and individuals to utilize large model technology.
AiBase Highlights:
🔑 360 Group opens the 360 Security Large Model for free, aiming to popularize large model technology and promote technological progress in the security industry.
💡 Free provision of large model standard functions to users who purchase 360 standard products, enhancing product functionality.
🚀 360 Group has successfully developed a self-developed general-purpose large model with thousands of parameters—360 Intelligence Brain, demonstrating its influence and technical strength in the market.
12. Lei Jun Says If Not for Car Manufacturing, He Would Be Working on AI: Took a Bunch of AI Courses Over 30 Years Ago
Lei Jun revealed in a recent interview that if Xiaomi had not entered the automotive manufacturing sector, the company might have focused more on the artificial intelligence field. Despite investing in car manufacturing, Xiaomi has not lagged behind in AI, having fully invested in the AI field as early as 2016. Lei Jun himself has a deep understanding of AI, and although his main focus in the past two years has been on the car project, he is still actively involved in AI-related work.
AiBase Highlights:
🚗 Xiaomi might have focused more on the artificial intelligence field if it had not entered the automotive manufacturing sector.
💡 Xiaomi has not lagged behind in the artificial intelligence field, having fully invested in AI as early as 2016.
💼 Lei Jun has a deep understanding of AI, and although his main focus in the past two years has been on the car project, he is still actively involved in AI-related work.
