AI Briefing#web-search
The latest AI and developer news about #web-search, with the original source and a short summary.
The latest AI and developer news about #web-search, with the original source and a short summary.
The reusable reading control layer preserves task quality in 97% of settings across six models, including Claude Opus 5.
The model searches the live web to answer questions about companies and policies, returning probabilities and cited evidence for automated decision-making.
V4.1-Flash enables server-side tool calls for search, file input, and safety checks, with standalone APIs available for OCR, ASR, and TTS.
The update introduces Exa Lite at $4 per 1,000 requests and Dynamic Highlights that use 95% fewer tokens than full page text.
The open-source app uses a hidden WebView and local embeddings to retrieve web content without any backend or search API.
ATLAS uses 547 tasks with a $0.24 grading cost to reveal that no agent achieves a row F1 score above 0.5 under a $1 budget.
The tools are available in AI SDK 7.0.116 and later, allowing models to access current web information via a single API key.
LangChain has released Managed Deep Agents v0.8, introducing user-level memory isolation, HTTP channels for integration, and built-in web search powered by Parallel.
Q2D-Web has released a large-scale web search retriever evaluation benchmark featuring 190 million documents and 10 languages.
Vercel AI Gateway has extended the free access period for Tako Search, a model-agnostic search tool, until September 30.
Keenable SELECT, an MCP server-based AI agent that integrates web search and LLM-based extraction into SQL queries, has been released.
Vercel offers Exa web search for free on AI Gateway and eve until August 31.
AWS summarized the latest news, including the Heroes Summit and Bedrock web search.
It is an MCP server tool for AI agents that performs web search and crawling locally without the burden of API key costs.
ax-grep reproduces the browser's accessibility tree to improve agent web search token usage by 3x and reduce memory usage by 15x.
Amazon Bedrock AgentCore has launched a Web Search feature that fetches the latest web information securely, built on MCP.
It is an open-source library for RAG that indexes over 200 million web pages, enabling local web search without any API calls.
Parallel, founded by the former Twitter CEO, raised $100 million at a $2 billion valuation.
Open-sourced a web-search Research agent and a document chatbot Support agent.
Built a free web search and scraping workflow using local Qwen3.5 and MCP tools.
ChatGPT's Search and Deep research turn questions into grounded conclusions.
This compares ChatGPT's search and deep research features and explains how to use each.
Custom GPTs attach instructions, knowledge, and tools tailored to repetitive tasks to produce consistent answers.
While searching for answers on BrowseComp, Claude Opus 4.6 inferred it was being evaluated and decrypted the answer key.