Google is replacing Gemini Gems with Skills. Rolling out in October 2026, Skills let you apply multiple specialized AI personas in a single conversation β something Gems never allowed. Here's a breakdown of the key differences, the transition timeline, and practical migration tips.
Google released Gemini 4 Argon on October 13, 2026. The model pushes output tokens to 1 million, targets complex coding and agentic tasks, and scored 77.9% on DeepSWE v1.1 β ahead of GPT-6 Astra (74.1%) and Claude Opus 5.5 (74.2%).
Claude Code Review reached general availability on October 15 for Team and Enterprise plans, enabling automatic PR reviews with Quick, Standard, or Deep levels. Google embedded Gemini into Android Messages to detect scam texts in real time using Private AI Compute. Anthropic launched Claude Haiku 5.5 β its fastest and most affordable small model yet, with a 1M-token context window.
Google DeepMind's EmbeddingGemma 2, released on October 6, 2026, is an open-source multimodal embedding model that maps text, code, images, video, and audio into a single 768-dimensional space. With 740M parameters running entirely on-device, it enables AI-powered search over student notes, lecture videos, and audio recordings β all without an internet connection.
Google DeepMind's Gemini 3.8 Live Extended Thinking, released to general availability on September 15, 2026, is an audio-to-audio model that performs deep reasoning during live voice streaming. It ranked #1 on Artificial Analysis' Speech-to-Speech leaderboard, and its asynchronous function calls keep audio flowing even while complex tools run in the background.