news - News

Ai
blog_img

NousResearch Hermes Agent v2026.4.30 Introduces Autonomous Curator and Faster Startup

Hermes Agent v2026.4.30 adds the autonomous Curator, new inference providers, bundled integrations, and major startup performance gains focused on self-improving AI operations.

Ai
blog_img

OpenClaw v2026.4.29 Strengthens Messaging, Memory, and Runtime Reliability

OpenClaw v2026.4.29 introduces active-run steering, people-aware memory wiki improvements, broader model provider coverage, stronger gateway reliability, channel delivery fixes, and security-focused operational hardening.

Ai
blog_img

CopilotKit v1.56.5 Improves Runtime Streaming, LangGraph State Handling and React Chat Stability

CopilotKit v1.56.5 adds LangGraph state injection and delivers fixes for runtime stream conversion, reasoning lifecycle handling, Express compatibility and React chat surfaces.

Ai
blog_img

LangGraph 1.2.0a2 Adds Node-Level Error Handlers and Retry Fixes

LangGraph 1.2.0a2 introduces node-level error handlers, makes NodeTimeoutError retryable by default, and improves StreamChannel projection ordering.

Ai
blog_img

crewAI 1.14.4 Expands Azure, Vertex AI and Research Tooling

crewAI 1.14.4 adds Azure OpenAI Responses API support, new research integrations, improved persistence options, and a broad set of reliability fixes for agent workflows.

Ai
blog_img

LangChain 1.2.17 Adds New Human-in-the-Loop Respond Decision

LangChain 1.2.17 is a focused update that introduces a new `respond` decision in HITL middleware, improving control over human-in-the-loop AI workflows.

Ai
blog_img

Caveman v1.7.0 Adds Usage Receipts, Smarter Installation, and Leaner AI Agent Handoffs

Caveman v1.7.0 introduces token receipt tracking, a smarter multi-agent installer, lighter subagent handoffs, MCP tool-description shrinking, and a key macOS detection fix.

Ai
blog_img

ZeroClaw v0.7.4 Strengthens AI Agent Onboarding, Internationalization and Session Control

ZeroClaw Release v0.7.4 introduces a Matrix rewrite, Mozilla Fluent internationalization, a rebuilt onboarding flow, restored WeChat iLink Bot support, and stronger session controls.

Ai
blog_img

MemPalace v3.3.4 Adds Local AI Setup and Cross-Project Memory Linking

MemPalace v3.3.4 introduces local LLM-powered setup, cross-project topic linking, one-step onboarding and ingest, and a major storage fix that cuts disk usage dramatically.

Ai
blog_img

LobeHub v2.1.56 Adds Briefs Schema Fields for Trigger and Metadata

LobeHub v2.1.56 introduces a small but important database migration, adding trigger and metadata fields to the briefs table plus a new index to support safer automation and richer business state.