Search AI Products and News

The Meta AI research team has made another breakthrough in the artificial intelligence field, officially releasing the new video understanding model V-JEPA2 (Video Joint Embedding Predictive Architecture2) on June 11, 2025. Led by Meta's chief AI scientist Yann LeCun, this model has opened up new possibilities in video understanding and physical world modeling with its innovative self-supervised learning technology and zero-shot robot control capabilities.

Global leading AI voice technology company, ElevenLabs, has officially released its latest text-to-speech model, Eleven v3 (Alpha version). It is considered the most expressive AI voice model to date. This breakthrough not only enhances the naturalness and emotional expression of speech synthesis but also provides content creators and developers with more powerful tools to assist in video, audiobook, and multimedia tool development. Technical Breakthrough: More Natural Dialogue and Emotional Expression. Eleven v3 introduces a new

With the rapid development of artificial intelligence technology, the text-to-speech (TTS) field has reached a new milestone. On June 5, 2025, ElevenLabs officially launched its latest text-to-speech model, Eleven v3 (Alpha version), known as the 'Strongest' TTS model. This model can not only convert text into natural and fluent speech but also simulate tone changes and non-verbal expressions in real conversations through precise emotional control and multi-language support, providing creators and developers with unprecedented voice generation capabilities.

According to the latest information, Midjourney's video function is about to be launched this month, while the development of V7.1 and V8 models is also accelerating. This article will provide you with a detailed analysis of Midjourney's latest developments, including breakthroughs in video functions, server upgrades, style references (sref) optimization, as well as future model plans, giving you an overview of the cutting-edge progress in AI image generation. The video function is ready to launch, and the rating party will start next week.

Recently, AIbase obtained the latest information from social media platforms and learned about Shisa.AI, a HuggingFace model provider specializing in Japanese fine-tuning. Its newly released bilingual Japanese-English model has attracted significant attention in the industry. This article will provide you with an in-depth analysis of Shisa.AI's latest achievements and its breakthrough advancements in the field of Japanese AI. Shisa V2405B: The birth of Japan's strongest open source model According to AIbase's understanding, Shisa.AI has recently released a Llama3.1-based S

Recently, the Speech Team of Qwen Laboratory has made a milestone achievement in the field of spatial audio generation and launched the OmniAudio technology. This technology can directly generate FOA (First-order Ambisonics) audio from 360-degree videos, bringing new possibilities to virtual reality and immersive entertainment. Spatial audio, as a technology that simulates real auditory environments, can enhance immersive experiences. However, existing technologies are mostly based on fixed perspective videos and underutilize the spatial information of 360-degree panoramic videos. Traditional video-to-audio generation

Recently, a new evaluation benchmark - RBench-V, specifically designed to test the visual reasoning capabilities of multi-modal large models, was released by research teams from Tsinghua University, Tencent HUNYUAN, Stanford University, and Carnegie Mellon University. The introduction of this benchmark aims to fill the gap in the current evaluation system regarding the model's visual output capabilities, allowing for a more comprehensive understanding of existing model performance. The RBench-V benchmark consists of 803 questions covering multiple fields, including geometry and graph theory, mechanics and electromagnetism, multi-target recognition, and path planning.

Recently, the DeepSeek team released a technical paper about their latest model, DeepSeek-V3, focusing on the challenges of scaling in large AI model training and related considerations of hardware architecture. This 14-page paper not only summarizes the experiences and lessons learned during the development of V3 by DeepSeek but also provides deep insights for future hardware design. Notably, CEO of DeepSeek, Liang Wenfeng, also participated in writing the paper.

PixVerse officially releases the V4.5 video model, adding over 20 movie-grade lens controls and multi-image reference features, as well as smoother processing capabilities for complex actions (https://pixverse.ai). This update significantly improves video generation quality and creative freedom, consolidating PixVerse’s leading position in the AI video generation field. AIbase observes that the release of V4.5 has quickly sparked global creator discussions, hailed as a milestone in movie-grade AI video creation. Core features: professional.

On April 25th, Razer officially launched its new Pro Click V2 series of ergonomic mice, including the standard Pro Click V2 and the new vertical Pro Click V2 Vertical Edition, starting at 799 yuan. These mice not only combine comfort and functionality in their design but also prioritize user experience. The Pro Click V2 features a classic 6-button design with a central scroll wheel...

On April 24th, Kunlun Wanwei announced the official open-sourcing of its multimodal reasoning model, Skywork-R1V2.0 (hereinafter referred to as R1V2.0). This upgraded version demonstrates significant improvements in both visual and text reasoning capabilities, particularly excelling in deep reasoning for challenging science problems in the College Entrance Examination and general task scenarios. It is considered the currently most balanced open-source multimodal model, equally adept at visual and text reasoning.