Anthropic doubled Claude Code five-hour limits after a SpaceX Colossus 1 compute deal. The AI coding agent race is shifting from model quality to infrastructure, quotas, and long-running reliability.
Claude Code 2.1.139 adds /goal, a session-level completion loop where a separate evaluator decides when an agent has actually finished.
Anthropic released 10 Claude agent templates for finance and insurance, showing how regulated industries may package agents around workflows, data, approval, and audit.
JetBrains AI Pulse says Claude Code reached 91% CSAT and an NPS of 54 while GitHub Copilot growth stalled, pushing AI coding toward a best-of-breed market.
MiniMax M2.7 uses a self-evolution loop around OpenClaw, activates only 10B of 230B parameters, and challenges premium coding models on price, benchmarks, and licensing.
Anthropic is limiting Claude Mythos Preview to Project Glasswing partners after reporting large jumps in autonomous vulnerability discovery, exploit chaining, and cyber safety risk.
China-based Z.ai released GLM-5.1 under MIT terms and topped SWE-Bench Pro with a 744B MoE coding model, sharpening the open-source versus closed-model split.
Anthropic added Monitor to Claude Code v2.1.98, letting Claude watch background command output and react to logs, CI status, and file changes while a coding session continues.
Alibaba released Qwen3.6-Plus for agentic coding with a 1M-token context window, free preview access, and a closed-source strategy that changes how builders should read Qwen.