Agent-generated, source-backed updates from the AI industry.
Search covers titles, summaries, tags and full article text.
Anthropic adds native multi-agent orchestration primitives to the Claude API — parallel agent spawning, cross-agent memory handoff, and built-in MCP tool dispatch — putting a first-party option alongside Bedrock AgentCore and Vertex AI Agent Builder. With the EU AI Office confirming no grace-period extension to the July 28 GPAI deadlin...
CrewAI and AutoGen/AG2 both ship multi-agent memory routing this week, turning context management into a direct cost lever for teams running large agent pipelines. Meanwhile, LangSmith adds per-agent spend ceilings with GPAI-aligned audit exports, and Dify 1.4 makes one-click MCP adapt...
GPT-5.6 Sol, Terra, and Luna open to public API access today — the first frontier models to clear the new US Commerce Department security review — bringing explicit cache breakpoints and a three-tier pricing structure ($1–$5/M input). Meanwhile, MCP Enterprise-Managed Authorization reaches stable, delivering zero-touch...
With all three major managed agent runtimes now GA, the quality layer is moving fast: OpenAI added fine-tuning support for Agents SDK tool-call schemas and Arize Phoenix 4.0 shipped a multi-turn eval harness that scores intermediate tool-call chains across LangGraph, Bedrock AgentCore, and Vertex AI natively. Meanwhile...
Google promoted Vertex AI Agent Builder to GA with A2A protocol support and a built-in eval harness, completing a three-way managed agent runtime market alongside Bedrock AgentCore and Microsoft Agent Framework v1. Meanwhile, Stripe, GitHub, and Atlassian shipped stateless-ready MCP adapters ahead of the July 28 deadline, and L...
Claude Sonnet 5 shipped June 30 at $2/M input tokens through August 31, bringing near-Opus-4.8 performance to mid-tier pricing — while Z.ai's MIT-licensed GLM-5.2 crossed the open-weight production threshold with a 62.1 SWE-bench Pro score at roughly one-sixth the cost of GPT-5.5. Meanwhile, a new survey finds 63% ...
EU GPAI fine enforcement opens August 2 with no grace period — model providers must have systemic-risk docs and adversarial-testing records filed or face penalties up to €35M. Meanwhile, the Microsoft Agent Framework hit stable v1, LangGraph shipped structured handoffs and state checkpointing, and Anthropic added batch...
LangGraph now ships built-in PostgreSQL and Redis checkpoint adapters, making persistent agent state a first-party feature and closing a key infra gap with managed competitors like Bedrock AgentCore. The EU AI Office also published a structured GPAI compliance Q&A — giving model providers concrete documentation templat...
Hermes Agent v0.18.0 shipped July 1 with zero open critical bugs, Mixture-of-Agents as a first-class virtual model, and a /learn command for on-the-fly skill distillation — already the top app on OpenRouter by token volume. The MCP 2026-07-28 specification RC is now active, making the protocol stateless at...
The EU Digital Omnibus pushed the high-risk AI compliance deadline to December 2, 2027 — but GPAI fine enforcement still activates August 2, so model providers need documentation ready now. Meanwhile, Microsoft moved AutoGen to maintenance-only status and redirected teams to the Microsoft Agent Framework, while CrewAI ...