Xiaomi unveils its MiMo-V2 AI models, including the 1T-parameter MiMo-V2-Pro, codenamed Hunter Alpha, which Xiaomi says benchmarks close to GPT-5.2 and Opus 4.6
Led by Fuli Luo, a veteran of the disruptive DeepSeek R1 project, the release represents what Luo characterizes as a “quiet ambush” on the global frontier.
Sources: Tencent is developing a top-secret AI agent for WeChat, and has tested using models from Zhipu, Alibaba, and DeepSeek, to compete with Qwen and Doubao
Tencent Holdings is secretly building a new AI agent for its hugely popular WeChat messaging app, in hopes of leapfrogging rivals …
Sources: DeepSeek plans to release its multimodal model V4 next week and worked with Huawei and Chinese AI chipmaker Cambricon to optimize V4 for their products
ByteDance's new AI video generation model Seedance 2.0 goes viral in China, with one state-backed newspaper saying it is bigger than DeepSeek's “Sputnik moment”
all production-ready. 15-sec multi-shot output with dual-channel audio. Film, advertising, gaming content costs about to crater. [video]@lentils80:Seedance 2.0 officially launches https://seed.bytedan...
ByteDance's new AI video generation model Seedance 2.0 goes viral in China, with one state-backed newspaper saying it is bigger than DeepSeek's “Sputnik moment”
ByteDance's new video-generating artificial intelligence model has already impressed the likes of Elon Musk and gone viral in China …
Multiple responses from DeepSeek's namesake chatbot confirm that the startup has expanded the context window of its flagship AI model from 128K tokens to 1M+
The upgrade will allow DeepSeek's AI model to remember and process more information in a single conversation or task
DeepSeek launches DeepSeek-OCR 2, an upgraded optical character recognition model that replaces OpenAI-developed CLIP framework with Alibaba's Qwen2-0.5b
Ben Jiang /South China Morning Post:
DeepSeek says its R1 update can perform mathematics, programming, and general logic better than the previous version, and comes close to o3 and Gemini 2.5 Pro
The Chinese startup DeepSeek said Thursday that its upgraded artificial-intelligence model can perform mathematics, programming …
Anthropic's Claude 3.7 Sonnet reportedly cost “a few tens of millions of dollars” to train, similar to Claude 3.5 and cheaper than GPT-4, which cost over $100M
“Assuming Claude 3.7 Sonnet indeed cost just ‘a few tens of millions of dollars’ to train, not factoring in related expenses, it's a sign of how relatively cheap it's becoming to release state-of-the-...
Tencent is testing DeepSeek for search in its messaging app Weixin, while Baidu plans to fully connect its search engine to DeepSeek and its own LLM Ernie
Tencent said on Sunday its (0700.HK) Weixin messaging app, China's largest, is allowing some users to search via DeepSeek's artificial …