AI Tools news and analysis on Web Pulse: 32711 curated articles tracking the latest AI Tools developments, tools, and research, updated continuously from vetted sources.
An AI agent discovered a code execution vulnerability in IDA Pro 9.3 for aarch64, a reverse engineering tool used by security researchers. The flaw allows attackers to inject arbitrary Clang arguments through crafted `.i…
Code Arena has introduced seven new category-specific leaderboards for web development tasks, including Reference-Based Design, Brand and Marketing Websites, and Simulations, to better measure AI model performance across…
Ably has released `@ably/ai-transport`, a new SDK that introduces branching conversation trees for AI interactions, allowing users to regenerate replies, edit messages, and navigate between sibling branches without losin…
Adding manpower to a late software project makes it later, and replacing specialized engineers with AI tools to fill gaps in a late project similarly delays it. Tech companies are laying off domain experts, citing AI as …
The 543rd edition of the Food for Agile Thought newsletter, shared with 35,597 peers, features John Cutler's warning that AI accelerates both broken and good practices equally. Guy Champniss adds psychological insights, …
OpenAI launched three new streaming audio models in its Realtime API: GPT-Realtime-2, GPT-Realtime-Translate, and GPT-Realtime-Whisper. GPT-Realtime-2 brings GPT-5-class reasoning to real-time voice agents with improved …
Supabase has launched an official app within ChatGPT, enabling users to manage their database infrastructure through natural language commands. The integration provides 29 tools for executing SQL queries, modifying schem…
OpenAI's Codex CLI and Anthropic's Claude Code are terminal-based AI coding agents built on different models, with Claude Code excelling at deep reasoning for complex refactoring and ambiguous requirements while Codex CL…
XAI's text-to-speech models outperformed OpenAI's in a blind comparison of six real-world scenarios, winning 20 out of 30 audio pairs and tying in six. In a separate realtime voice agent test for a hotel concierge, both …
Zed Industries released Zeta2.1, an updated edit prediction model that emits 3x fewer output tokens than its predecessor, reducing response times by up to 50 milliseconds and requiring 30% fewer servers to handle the sam…
Google's Nano Banana Pro now matches GPT Image 2 on clean English typography and delivers a more polished editorial layout, but GPT Image 2 still leads in multilingual text rendering, dense scene accuracy, and edit fidel…
Glama lists over 22,000 MCP servers, PulseMCP tracks 14,000+, and mcp.so and other registries add thousands more, but a handful of well-chosen servers composed thoughtfully outperform a full stack for startup teams. The …
Modular announced that traditional HTTP-era load balancing algorithms like round-robin, consistent hashing, and least-connections are inadequate for large language model inference because GPU pods are stateful, specializ…
Netflix released void-model on Hugging Face, and a developer used the Goose CLI agent with Together's dedicated containers skill to deploy the model for inference on release day with a single prompt. The agent automatica…
A developer frustrated with the fragmented local AI model ecosystem argues that running models locally for coding agents remains unnecessarily difficult compared to using hosted APIs, citing missing features like tool pa…
Mozilla AI released Octonous into open beta, making the AI assistant available to all users without a waitlist. The company incorporated feedback from its closed beta, shipping improvements including expanded integration…
A developer built a structured-retrieval discovery agent for the educational movie-discovery project Kino, using LangGraph to enable prompt-driven catalog searches. The system interprets natural language queries like "Di…
Developers seeking alternatives to Windsurf in 2026 are primarily motivated by pricing predictability, local-first workflows, and access to real agent engines like Claude Code or Codex. Nimbalyst leads as the top alterna…
OpenAI's GPT-5.5 Codex model was tested at four reasoning effort levels — low, medium, high, and xhigh — on 26 real-world coding tasks from the open-source GraphQL-go-tools repository. Low and medium tied on test pass ra…
A developer has created a CLAUDE.md snippet that enforces a "graphify-first, grep-last" workflow for all cross-module exploration, requiring Claude Code to use graphify's query, path, explain, and wiki tools before falli…