Track what changed across AI agents

Anthropic

Claude Opus 4.7 is described as Anthropic's most capable vision model, powering the new Claude Design product. This release enables advanced visual work including designs, prototypes, slides, and more through a collaborative interface. The model supports importing from various file formats, inline editing, design system integration, and handoff to Claude Code.

Suno AI

Suno has launched Suno Studio, described as the world's first generative audio workstation that integrates AI-powered generation at the heart of music creation. It features multitrack timeline editing, stem generation (vocals, drums, synths), BPM control, volume and pitch adjustment, and the ability to export stems as audio and MIDI files. It is available on desktop for Premier users.

OpenAI

GPT-5.5 is described as OpenAI's smartest model yet, offering faster performance and increased capabilities. It is built for complex tasks including coding, research, and data analysis across tools.

Cursor AI

Cursor introduced multitasking with async subagents via /multitask command to parallelize requests, improved worktrees experience in the Agents Window for running isolated tasks across branches, and multi-root workspace support allowing a single agent session to target multiple folders for cross-repo changes.

Cursor AI

This release introduces multitasking via async subagents using the /multitask command, allowing parallel processing of requests. It also adds improved worktrees support in the Agents Window for running isolated tasks across different branches, and introduces multi-root workspaces enabling a single agent session to make cross-repo changes spanning frontend, backend, and shared libraries.

Cursor AI

Cursor 3.1 introduces interactive Canvases that allow the AI to respond with visualizations including dashboards and custom interfaces. The Agents Window received a tiled layout for running multiple agents in parallel, and voice input was upgraded with batch STT for higher-quality speech-to-text transcription.

Cursor AI

Bugbot can now learn from feedback on pull requests and turn those signals into learned rules that improve future reviews, achieving a 78% resolution rate. MCP server support was added for additional context during code reviews on Teams and Enterprise plans. Additional improvements include a 'Fix All' action, redesigned settings, and improved Bugbot Autofix reliability.

Suno AI

Suno v5.5 introduces new features focused on expressiveness and personalization. The update includes Voices, Custom models, and a 'My Taste' feature, allowing users greater creative control over their music generation.

Replit

Agent 4 was launched by Replit, representing a significant new version of the Replit Agent. It supports a unified project experience on an Infinite Canvas, allowing builders to create landing pages, web apps, and mobile-native versions all within one project. The launch was celebrated with a behind-the-scenes stream featuring the Replit team.

Vercel

Added support for .riv file uploads in chat and projects, and added element screenshot attachment for design mode element selection. Improved git sync workflow with simplified PR menu, and fixed several bugs including design mode sidebar flash, layout shifts in Data Viewer, and turborepo interactive task startup issues.

Vercel

v0 received multiple updates in March 2026 including support for .riv file uploads, element screenshot attachments for design mode, improved git sync workflow, and image generation capability added to the Plan agent. Next.js was updated to version 16.2.0, and various bug fixes were applied including form submission issues during streaming and binary asset preservation in git-backed zip imports. Additional improvements include team skills linking to design systems, automatic PR labeling with 'v0', and extended monthly credit expiry from 30 to 65 days.

Mistral AI

Mistral AI introduced Mistral Small 4, a new model version. Additionally, Voxtral TTS was released as a frontier, open-weights text-to-speech model that is fast, instantly adaptable, and produces lifelike speech for voice agents. Mistral also introduced Forge, a system for enterprises to build frontier-grade AI models grounded in proprietary knowledge.