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.
A punishing technology selloff erased billions in market value in the week ending July 17, with the technology sector tumbling 5.9% and the iShares Semiconductor ETF sinking 19% month to date, as inve…
A University of Cambridge field study based on interviews with former Boko Haram and ISWAP members reveals that these groups have set up dedicated internal units to systematically use commercial AI ch…
A Twitter user investigating potential distillation of proprietary AI models by DeepSeek found that DeepSeek V4 Pro, accessed via OpenCode and the official DeepSeek API, produces outputs nearly identi…
Waylou ACE CLI, an open-source AI coding assistant forked from Google Gemini CLI, adds multi-provider support for Gemini, OpenAI, Anthropic, Ollama, and DeepSeek models through a single terminal inter…
Moonshot's Kimi K3 is the first Chinese open-source AI model to match the performance of top American frontier models, landing on par with Anthropic's Mythos/Fable and OpenAI's GPT-5.6 across benchmar…
Zhipu, the Tsinghua University spinout behind the GLM AI models, has seen its stock surge nearly 1,500 percent since its January IPO at HK$116.20, reaching a peak market cap above $100 billion, despit…
Chinese company Moonshot AI released its open source Kimi K3 model, which independent analyses show is competitive with frontier proprietary models, spooking Wall Street and causing the Nasdaq to drop…
Z.ai, the Tsinghua University spinout co-founded by Tang Jie and CEO Zhang Peng, reached a July sales pace that would annualize to $1 billion, according to Bloomberg, which would make it the first Chi…
A developer deleted a line from their bio boasting about running hand-built inference engines for open-source models like Qwen and DeepSeek, realizing it was a 'demo' rather than a true flagship. The …
An independent test of DeepSeek-V4-Pro found sharply different behavior depending on whether prompts contained cyber or biology content, raising new questions about whether some requests were being an…
Chinese AI firms Moonshot AI and MiniMax unveiled powerful new models at the World AI Conference in Shanghai on July 17, triggering a broad sell-off in US tech stocks. The Nasdaq dropped 1.4%, the S&P…
A developer running a small training platform found that a bug in the evaluation harness, not the models, was preventing five open-source models from passing a 54-task exam. The same pattern repeated …
DeepSeek trained its flagship V3 model for about $5.6 million, a fraction of the $50 million to $100 million OpenAI spent on GPT-4, and the Chinese lab operating under U.S. chip export controls has co…
Moonshot AI released Kimi K3, a 2.8 trillion parameter open-weight model, on July 16, and within a day it topped the Frontend Code Arena leaderboard with a score of 1679, beating Anthropic's Claude Fa…
TSMC reported blowout Q2 earnings with revenue jumping 36% year over year and net income soaring 77.4%, driven by surging AI chip demand, while IBM warned Q2 revenue and earnings fell below expectatio…
Beijing-based Moonshot AI launched Kimi K3 on July 16, a 2.8 trillion parameter Mixture-of-Experts model that beats Claude Opus 4.8 and GPT-5.5 on coding and agent benchmarks, making it the largest op…
Chinese open-weight AI models now occupy the top five spots on OpenRouter by weekly token usage, threatening to turn America's premium models into expensive niche products. Moonshot AI's Kimi K3, rele…
Moonshot AI, a Beijing-based startup founded by Yang Zhilin, unveiled Kimi K3 on July 16-17, 2026, as the world's largest open-weight AI model with 2.8 trillion parameters, outperforming OpenAI's GPT-…
OpenAI released the GPT-5.6 model family last week, which comes in three sizes each with roughly five or six reasoning-effort settings, according to Sebastian Raschka. The article explains how reasoni…
Mozilla's inaugural State of Open Source AI report finds that 79% of developers use open models but only 51% have deployed one to production, revealing a 28-point deployment gap. The report, based on …