Zhipu AI introduces GLM-5.3-Flash, a lightweight LLM with open weights and commercial use rights.
The latest Qwen3-2507 series is now open-sourced, featuring 235B/30B/4B MoE and dense models.
NVIDIA executives address funding concerns and advance chip supply and AI ecosystem initiatives.
NVIDIA reports $10 billion quarterly revenue, marking a historic milestone.
MSI's XpertStation WS300 delivers desktop-scale DGX-class performance for enterprise AI workloads.
OpenAI's Jalapeño inference chip surfaces with benchmark data, revealing accelerating hardware divergence across Prefill and Decode phases.
The story of my first accepted, merged, and published security advisory — a critical RCE bypass in qwed-mcp — and what the earlier asteval rejection taught me.
A California federal judge ruled the Pentagon's blacklist of Anthropic constitutes unlawful First Amendment retaliation.
DeepMind partners with Singapore AISI, OpenMined, AVERI, and MLCommons to test Gemini Flash Lite in a privacy-preserving environment.
Zhipu launches new multimodal model with native visual-text fusion and Agent task optimization.
