Much like Anthropic’s Computer Use, ByteDance’s new UI-TARS understands ... consistently beating out OpenAI’s GPT-4o, Claude and Google’s Gemini. “Through iterative training and ...

ByteDance’s Doubao 1.5 Pro is a cost-effective ... s model and more than 200 times less expensive than OpenAI’s GPT-4o. Doubao also boasts a 32k + 256k context and utilises a sparse Mixture ...

Amid the USAID shutdown controversy, Indian-origin Perplexity AI CEO Aravind Srinivas took a bold jab at billionaire Elon ...

The Chinese owner of TikTok rolled out the new software targeting overseas Chinese coders after a reprieve from the Trump administration.

The artificial intelligence landscape is experiencing a seismic shift, with Chinese technology companies at the forefront of ...

Researchers at ByteDance, TikTok's parent company, showcased an AI model designed to generate full-body deepfake videos from one image and audio — and the results are scarily impressive.

The use of motion signals is a novel technique, which the company is calling omni-conditions training. With this, the AI model is trained on different modalities, including text, image, audio, and ...

Explore TikTok’s new TRAE AI coding assistant is helping software development with GPT-4, multimodal inputs, and real-time ...

Alibaba unveils Qwen 2.5 AI model, claiming it outperforms GPT-4o and DeepSeek-V3 as China’s AI race intensifies.

ByteDance’s OmniHuman-1 generates lifelike human videos from a single image and audio. Discover its key features and compare it with Sora and Veo 2.