Large Language Models

On July 21, 2026, Google announced three new Gemini models at once: Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, and the defense-focused Gemini 3.5 Flash Cyber. The theme of the announcement is clear: not higher scores, but getting the same job done with fewer tokens. As agentic workflows move into production, the bill is driven less by how smart a model is and more by how many tokens that intelligence burns. Google aimed straight at that. Here are the details. ...

TL;DR: Kimi K3 in 30 Seconds Kimi K3 is the new AI model Chinese company Moonshot AI announced on July 16, 2026; at 2.8 trillion parameters, it is the largest open source model released so far. In a single 48-hour run it designed a working chip, finished an astrophysics study in 2 hours instead of 2 weeks, and cut a teaser video from 56 raw clips. It reads 1 million tokens at once: roughly like reading the entire Lord of the Rings trilogy in one sitting and remembering all of it. On coding it plays in the same league as the strongest Claude and ChatGPT models, and beats them on some tests. It falls behind on hard knowledge questions. You can try it in the Kimi app and on kimi.com. The model files go public on July 27, 2026. Ask an AI to design a chip for you and what happens? It probably writes you a nice article about how chips are designed. ...

TL;DR: GPT-Red in 30 Seconds GPT-Red is an automated red-teaming model OpenAI trained to break its own models; it is not public and never will be. It was trained with self-play: attacker GPT-Red and defender models played against each other, both getting stronger over time. In scenarios where human red-teamers succeeded only 13% of the time, GPT-Red hit 84%. GPT-5.6 was trained against GPT-Red, making it OpenAI’s most robust model against prompt injections (only a 0.05% failure rate). OpenAI just announced something unusual: not a model for users, but an “AI hacker” trained to break its own models. Meet GPT-Red! According to the research publication “Unlocking Self-Improvement for Robustness” released today, GPT-Red automates the work of human security testers (red teams) and does it far better than humans. ...

A powerful move has arrived on the price and efficiency front of the AI race! On July 8, 2026, xAI, Elon Musk’s AI company now operating under the SpaceX umbrella, announced Grok 4.5, its first major model release since going public. The model is also the fruit of a collaboration with the Cursor team, and its pitch is clear: not the highest benchmark score, but the most efficient way to get work done. ...

Talking to AI no longer feels like a walkie-talkie exchange, it finally feels like a real conversation! On July 8, 2026, OpenAI announced GPT-Live, its next-generation family of voice AI models. This is not just a better voice mode, it is an architectural shift that redefines the voice AI experience. So what makes GPT-Live so different? Let’s dive in. What Is GPT-Live? GPT-Live is OpenAI’s new voice model family built on a full-duplex architecture. Full-duplex means the model can listen and speak at the same time. With the previous Advanced Voice Mode, the model waited while you spoke and you waited while it spoke. GPT-Live removes that turn-taking constraint entirely. ...