AI News
Making Knowledge Distillation Cheap Enough to Run at Scale
The Kimi-K3 model requires 3TB of VRAM just to load, making it impossible to run on standard hardware.…
AI for Business
OpenAI acquires NextSlide to bring AI-generated presentations into ChatGPT
OpenAI has acquired NextSlide, the startup that converted prompts and documents into editable presentations. Founder Ahmed Beshry now…
AI for Business
These startups are chasing the next big thing in LLMs
MIT Technology Review’s What’s Next series examines industries, trends, and technologies to offer a first look at the…
AI Research & Science
AI for science needs reasoning, not just data
Albert Michelson wrote in 1903 that the facts of physical science had all been discovered. Stephen Hawking predicted…
AI News
Hidden text in a PDF is enough to steal sensitive data through Atlassian’s AI agent Rovo
Hidden text in a PDF is enough to steal sensitive data through Atlassian’s AI agent Rovo Key Points…
AI Music
ByteDance Seed Introduces SeedRealtime: a Native Audio-Visual Full-Duplex LLM That Watches, Listens and Speaks in One Model
ByteDance’s Seed team has launched SeedRealtime, a model that processes audio, video, and text within a single architecture.…
AI News
Quoting Claude Opus 5 system prompt
Claude Opus 5 system prompt documentation confirms that the model acknowledges the June 2026 suspension of access to…
AI News
NVIDIA Releases NemotronLabs VoiceChat 11B: An Open Full-Duplex Speech-to-Speech Model with ~450 ms Turn-Taking and Live Tool Calling
NVIDIA has released NemotronLabs VoiceChat 11B, an open 11B speech-to-speech model designed for real-time, full-duplex conversation. It processes…
AI News
SQLite compressed text-history prototypes
Simon Willison has released a prototype for storing text revision histories in SQLite that uses compression to keep…
AI News
GitHub Models is now retired
GitHub Models has been retired, a service that allowed developers to run large language model prompts directly within…