ai - News

Ai
blog_img

RAGFlow v0.25.6 Adds AI Browser Agent and Upgrades RAPTOR with Ψ-RAG

RAGFlow v0.25.6 introduces an AI Browser component for autonomous web navigation and Ψ-RAG with AHC mode for faster, broader semantic indexing.

Ai
blog_img

OpenClaw v2026.5.26 Boosts Gateway Performance, Makes Transcripts Core

OpenClaw's latest release cuts gateway latency and elevates transcripts to a first-class feature. Faster replies, reduced cache churn, and unified transcript pipelines for summaries, media provenance, and replay.

Ai
blog_img

specify-cli 0.8.15: Preset Updates and New Token Budget Extension

The latest specify-cli release updates the Fiction Book Writing preset, adds a Token Budget extension, and fixes hook dispatch.

Ai
blog_img

Mem0 v2.0.4 Adds Cascading Delete for Linked Memories

Mem0's latest SDK update introduces `delete_linked` parameter, allowing cascading deletion of superseded memories to prevent stale data from resurfacing.

Ai
blog_img

LangChain Perplexity 1.3.1: Dependency Update to PerplexityAI 0.34.1

The latest patch release for langchain-perplexity bumps the underlying perplexityai library to version 0.34.1. Here's what changed and why it matters.

Ai
blog_img

Gemini CLI v0.44.0 Tightens Type Safety with Strict Validation

Google's Gemini CLI v0.44.0 eliminates unsafe type suppressions, improving code reliability. A look at the refactor and why it matters.

Ai
blog_img

MoneyPrinterTurbo v1.2.8 Adds LiteLLM and Grok/xAI Support

MoneyPrinterTurbo v1.2.8 introduces LiteLLM provider for 100+ gateways, Grok/xAI integration, and WebUI custom audio upload. Improved Gemini TTS and Edge subtitle compatibility.

Ai
blog_img

OpenClaw v2026.5.27: Stricter Security and Codex Reliability Improvements

OpenClaw's latest update tightens security with new boundaries and improves Codex app-server reliability. Key changes include blocking side-effecting command wrappers and requiring admin authority for approvals.

Ai
blog_img

Docling v2.96.0 Introduces Threaded PDF Backend and JSON Transformers Fix

Docling v2.96.0 adds a threaded docling-parse PDF backend and accepts JSON for transformer model types, improving performance and flexibility.

Ai
blog_img

Meilisearch v1.45.1 Reverts Autobatch Deletions by Filter

Meilisearch's latest patch reverses a controversial autobatch feature. Here's what changed and why it matters.