DeepSeek is a Chinese AI research laboratory that has developed highly capable open-source language models including DeepSeek-V3 and DeepSeek-R1, notable for their efficiency and performance.
Chinese AI startup DeepSeek released its R1 and V3 reasoning models on January 27, 2025, wiping $589 billion from Nvidia's market cap as investors questioned the need for expensive hardware. Within ho…
Beijing-based Moonshot AI released Kimi K3 on Thursday, a 2.8-trillion-parameter open-source model that the company claims is the largest ever built, roughly 75% larger than DeepSeek's V4 Pro at 1.6 t…
Moonshot AI released Kimi K3, a 2.8 trillion-parameter open-weight model, on July 16 at the World Artificial Intelligence Conference in Shanghai, making it the largest open-source AI model to date. Ch…
A new study from the Oversight Board, an independent body funded by Meta, found that top AI models refuse political speech requests about repressive governments at more than twice the rate as requests…
China's Moonshot AI released Kimi K3, a massive new model that matches or beats top U.S. models like Anthropic's Fable 5 and OpenAI's GPT-5.6 Sol in coding tests while costing 40% less, erasing Americ…
The Oversight Board's first evaluation of large language models from Anthropic, DeepSeek, Google, Meta and OpenAI found that models are more than twice as likely to refuse to criticize repressive regi…
Chinese AI models now power 60% of US corporate AI use on the OpenRouter platform, with DeepSeek as the most popular model, according to data from the platform. The shift has geopolitical implications…
A Chinese luggage maker's stock-exchange filing implies DeepSeek is valued at around $52bn, the most precise figure yet for the Hangzhou AI startup. Anhui Korrun's subsidiary Ningbo Purun took a 0.826…
A developer built an OpenAI-compatible prompt encryption proxy in ~300 lines of Python using FastAPI and httpx, after their GPT prompts were stolen and resold. The proxy keeps the system prompt on the…
Chinese President Xi Jinping called for global cooperation in artificial intelligence development and governance at the World Artificial Intelligence Conference in Shanghai, opposing the overstretchin…
A user reports using DeepSeek for 80% of private AI work, citing high satisfaction, while criticizing enterprise Copilot for degraded quality, slow speed, and API errors, and noting that GPT-5.4 has d…
Chinese President Xi Jinping made his first appearance at the World Artificial Intelligence Conference in Shanghai on July 17, praising China's low-cost AI breakthroughs and calling for an open intern…
Chinese AI startup DeepSeek raised over $7.4 billion in its first external funding round, achieving a post-money valuation above $50 billion and becoming China's most valuable AI startup. Investors in…
Moonshot AI's Kimi K3, a 2.8 trillion parameter open-weight model, topped Arena.ai's Frontend Code leaderboard with 1,679 points, beating Claude Opus 4.8 and other frontier models, and is priced at $3…
Chinese President Xi Jinping made his debut at the World AI Conference in Shanghai on Friday, calling for a 'people-centred' and inclusive approach to AI development to prevent 'new historical injusti…
Chinese President Xi Jinping is gaining global influence as low-cost AI models from Chinese labs like DeepSeek and Alibaba close the technology gap with US rivals, with DeepSeek's R1 and V3 models cos…
An open-source AI agent called SAM that runs on a user's own machine is now available, offering a free, private team of AI agents that can research, remember, and take real action on the computer. The…
Beijing-based Moonshot AI on July 16 unveiled Kimi K3, the largest open-weight AI model with approximately 2.7 trillion parameters using a Mixture-of-Experts architecture, nearly doubling the 1.6 tril…
Treena, a new agentic coding environment designed for mobile phones, lets developers describe a task, approve a plan or let the agent run, test, and push to GitHub entirely from a phone without a lapt…
Chinese startup Moonshot AI released Kimi K3, a 2.7-trillion-parameter open-weight large language model, on July 16, claiming it performs competitively with Anthropic's Fable 5 and substantially outpe…