#LLM
Curated coverage on «LLM»: releases, research, benchmarks, open-source.

Anthropic Is Building a 'Tamagotchi' Into Claude: Leak Reveals New Chatbot Features
An accidental Anthropic update has exposed Claude's roadmap. The company plans gamification features, including a virtual pet meant to build an emotional bond with…

Nexus Raises $4.3M to Let Non-Programmers Build Enterprise AI Agents
Belgian startup Nexus has raised $4.3 million. Its platform lets non-technical staff deploy enterprise AI agents quickly, with no programming skills required.

Evolving AI Agents: A Step-by-Step Guide to the A-Evolve Framework
Learn how to build and configure an evolving OpenAI-based AI agent with the A-Evolve framework. A hands-on guide to memory, skills, and the agent workspace.

The Road to Autonomy: How One AI Agent Grew Into a Digital 'Village'

The Evolution of AI Assistants: Why Existing Frameworks Fall Short of the New MicroClaw

BNP Paribas Scales Up AI: Every Employee Gets a Digital Assistant

Google Translate's Live Translate Now Works on iOS

Gemini 3.1 Flash Live Released: A New Standard for Natural Voice AI
Gemini 3.1 Flash Live is here. Its developers sharply cut latency and improved accuracy, aiming for AI conversation that feels live and fluid.

Gemini 3.1 Flash Live: Audio AI Gets More Natural and More Reliable
Google's new Gemini 3.1 Flash Live model is now available across all of its products. The update aims to make voice AI interactions more natural and more reliable.

The Magic of Quantization: How to Run Giant Neural Networks on an Ordinary Laptop

OpenAI Shelves Adult AI Chatbot Project to Focus on Business Goals

SEO Is Evolving: How to Get Into ChatGPT and Perplexity Answers

Google unveils TurboQuant algorithm: AI model memory cut sixfold

How AI agents defend against prompt injection
ChatGPT has learned to resist manipulation: here is how AI agents block dangerous commands and protect user data.

Rakuten Cuts Incident Resolution Time in Half With AI
Japanese tech giant Rakuten deployed an AI coding agent and cut system recovery time by 50%.









