Ask by product,
task, or condition.

ChatGPT can connect to custom apps built with MCP, with workspace controls and plan-dependent app capabilities.

ChatGPT exposes web search as a conversation tool, including inside Projects when the workspace allows it.

ChatGPT agent can use its own computer to complete multi-step web tasks under user supervision.

ChatGPT executes Python for data analysis on Free, Plus, Pro, Business, Enterprise, and Edu, with Free on limited data-analysis quotas.

ChatGPT accepts file uploads on Free, Plus, Pro, Business, Enterprise, and Edu, with Free limited daily uploads and paid plans at higher ceilings.

ChatGPT analyzes uploaded PDFs on Free, Plus, Pro, Business, Enterprise, and Edu for extraction and synthesis tasks, with Free on limited upload and data-analysis quotas.

ChatGPT analyzes CSV, XLS, and XLSX spreadsheets on Free, Plus, Pro, Business, Enterprise, and Edu through data analysis, with Free on limited quotas; a separate Excel add-in covers additional workspace plans.

ChatGPT accepts static image inputs on Free, Plus, Pro, Business, Enterprise, and Edu across web and mobile, with Free on limited usage.

ChatGPT Images creates new images and edits uploaded or previously generated images; basic generation is documented across all tiers.

ChatGPT Projects retain chats, files, and instructions as project context, while broader memory availability depends on settings and workspace policy.

ChatGPT runs background scheduled and monitoring tasks on Plus, Pro, Business, Enterprise, and Edu, including one-off, recurring, and change-detection work the user can leave and manage later.

ChatGPT connects to third-party apps and plugins on Free, Plus, Pro, Business, Enterprise, and Edu; Free includes directory plugins/interactive apps, while workspace admins gate Business/Enterprise/Edu connectors.

ChatGPT is documented across web and desktop workflows, with dedicated desktop applications for macOS and Windows.

Claude supports MCP across Claude.ai, Claude Desktop, Claude Code, and the Messages API, with setup varying by surface.

Claude Research uses live web search and returns cited research on paid Claude plans.

Claude can control a browser (navigate, click, type, fill forms) on Pro, Max, Team, and Enterprise through Claude in Chrome from Desktop chat/Cowork/Code and the Chrome side panel.

Claude provides a sandboxed computing environment for code execution, file creation, and data analysis.

Claude accepts documents and images in chats and projects, subject to file-size, count, and context limits.

Claude analyzes PDF text and, for PDFs under the documented page threshold, visual elements such as charts and graphics.

Claude accepts XLSX when code execution is enabled and can analyze CSV, TSV, and spreadsheet data in its computing environment.

Claude accepts JPEG, PNG, GIF, and WebP image uploads and can interpret supported visual content.

Claude does not generate or edit photos or illustrations as an image-generation product; it is limited to image understanding plus HTML/SVG visuals in chat.

Claude can search and reuse context from previous conversations for paid-plan users where chat search and memory are available.

Claude runs background Dispatch agents on Pro and Max via Claude Cowork on Desktop, carrying out tasks after you leave when the desktop host stays online.

Claude supports remote connectors across its major surfaces and local desktop extensions on Claude Desktop and Claude Code.

Claude Desktop is documented for macOS, Windows, and Linux.

Perplexity supports custom remote connectors built with MCP in its enterprise connector system.

Perplexity combines live web search with uploaded or connected internal knowledge in supported search workflows.

Perplexity Computer can browse the web and automate browser actions on Pro, Max, Enterprise Pro, and Enterprise Max.

Perplexity Computer can create production-ready apps and run multi-step tool-backed workflows on Pro, Max, Enterprise Pro, and Enterprise Max.

Perplexity accepts local file uploads including text, code, PDFs, images, audio, and video on Free (limited), Pro, Max, Enterprise Pro, and Enterprise Max.

Perplexity analyzes uploaded PDFs on Free, Pro, Max, Enterprise Pro, and Enterprise Max via file attachments and knowledge workflows.

Perplexity analyzes spreadsheet data such as CSVs on Pro, Max, Enterprise Pro, and Enterprise Max, including Computer workflows that chart and summarize tabular files.

Perplexity accepts image attachments in individual queries, but its Internal Knowledge Search does not index PNG or JPEG files.

Perplexity generates images on Free, Pro, Max, Enterprise Pro, and Enterprise Max, with Free limited and higher tiers unlocking more high-quality generation.

Perplexity stores project-scoped memories, while Computer also documents persistent context across sessions.

Perplexity Computer runs background tasks and continuous monitoring on Pro, Max, Enterprise Pro, and Enterprise Max while the user is away.

Perplexity offers app, file, and custom remote connectors on Pro, Max, Enterprise Pro, and Enterprise Max, with enterprise admin controls for org-wide connector policy.

Perplexity Computer is available on desktop web and mobile; this evidence does not establish a native desktop application.

Codex supports MCP for connecting external tools and systems.

Codex includes a web-search tool for current external information.

The Codex app includes an in-app browser that Codex can control for localhost web development, with broader control described as evolving.

Codex runs software-engineering tasks in command-capable sandbox environments and is also available as a terminal CLI.

Codex executes code and tests inside isolated task environments while working on a repository.

Codex can inspect screenshots, technical diagrams, charts, and other visual surfaces during coding work.

OpenAI documents image generation and editing as available inside Codex.

Codex cloud tasks run independently in separate environments and can be delegated in parallel.

Codex connects to external systems through configured MCP servers and other available tools.

OpenAI offers a Codex desktop app with an integrated workspace and browser.

Codex operates across app, CLI, and IDE extension surfaces, with shared agent workflows.

Claude Code connects to Model Context Protocol (MCP) servers on Pro, Max, Team, Enterprise, and API (Console) access; the free Claude.ai plan does not include Claude Code.

Claude Code includes a built-in WebSearch tool that searches the web and returns result titles and URLs for current documentation and external information.

Claude Code can control a browser through the Claude in Chrome extension on Pro, Max, Team, and Enterprise for navigation, clicks, form fill, debugging, and data extraction.

Claude Code runs as an interactive terminal agent and can execute approved shell workflows.

Claude Code can inspect repositories, edit files, and run commands from its terminal agent loop.

Claude Code accepts static image inputs on Pro, Max, Team, Enterprise, and API for screenshot, mockup, and diagram analysis via drag-and-drop, path, or clipboard paste.

Claude Code runs background agent sessions that keep working without an attached terminal on Pro, Max, Team, and Enterprise, including Dispatch flows that continue while you are away.

Claude Code can use remote connectors and local desktop extensions through Anthropic’s connector architecture.

Claude Desktop can deep-link directly into a Code session on macOS, Windows, and Linux.

Claude Code supports IDE-integrated workflows in addition to its command-line interface.

Cursor connects to Model Context Protocol (MCP) servers on Hobby, Pro, Pro Plus, Ultra, Teams, and Enterprise through local and remote MCP configuration.

Cursor Agent includes web search and URL fetching among its research tools.

Cursor Agent can control a browser to navigate pages, click, type, fill forms, and take screenshots on Hobby, Pro, Pro Plus, Ultra, Teams, and Enterprise without installing external browser tools.

Cursor Agent can run commands in the editor terminal, and Cursor also provides a terminal-native CLI agent.

Cursor Agent can edit repository files and execute commands and tests through the terminal tool.

Cursor accepts static image inputs on Hobby, Pro, Pro Plus, Ultra, Teams, and Enterprise via Agent chat drag-and-drop or paste, and can attach screenshots on iPad.

Cursor Agent can generate images from text descriptions or reference images on Hobby, Pro, Pro Plus, Ultra, Teams, and Enterprise for mockups, assets, and diagrams.

Cursor Cloud Agents (formerly Background Agents) run delegated work in isolated cloud VMs launched from web, mobile, desktop, or Slack, and hand off results as pull requests.

Cursor’s MCP directory provides connectors for services including GitHub, Linear, Notion, Figma, and databases.

Cursor is an installed desktop code editor for macOS, Windows, and Linux.

Cursor is itself an AI-first IDE and exposes its agent tools inside the editor.

GitHub Copilot supports Model Context Protocol (MCP) client integrations on Free, Pro, Pro Plus, Business, and Enterprise, including Free individual seats not covered by the Business/Enterprise MCP policy gate.

GitHub Copilot Chat can search the web with Bing on Free, Pro, Pro Plus, Business, and Enterprise when Copilot access to Bing is enabled by the user or organization policy.

GitHub Copilot cloud agents can use an out-of-the-box Playwright MCP server against localhost.

GitHub Copilot CLI runs as an agent in the terminal and can perform approved command-line work.

GitHub Copilot agents can make code changes and use command execution in terminal, IDE, and cloud workflows.

GitHub Copilot accepts user file uploads in Chat on Free, Pro, Pro Plus, Business, and Enterprise, including images and PDFs attached to prompts.

GitHub Copilot analyzes uploaded PDF attachments on Free, Pro, Pro Plus, Business, and Enterprise in Copilot Chat when a supporting model is selected.

GitHub Copilot accepts static image inputs on Free, Pro, Pro Plus, Business, and Enterprise across github.com Chat, VS Code, and the Copilot CLI when a vision-capable model is selected.

GitHub Copilot cloud agent runs delegated coding work asynchronously and opens draft pull requests on Pro, Pro Plus, Business, and Enterprise, not on Copilot Free.

GitHub Copilot can connect to external services through configured MCP servers and organization-approved registries.

GitHub Copilot integrates with VS Code, Visual Studio, JetBrains IDEs, Xcode, and other supported development environments.

Gemini Code Assist uses configured Model Context Protocol (MCP) servers in agent mode on Standard and Enterprise in both VS Code and IntelliJ.

Gemini Code Assist agent mode can use Google Search responses in VS Code context.

Gemini Code Assist agent mode exposes Gemini CLI built-in tools, including terminal access, in VS Code.

Gemini Code Assist agent mode can write files and run tools as part of multi-step coding work.

Gemini Code Assist can connect to external services through MCP servers and RESTful API tools in agent mode.

Gemini Code Assist agent mode is available inside VS Code and IntelliJ-family IDEs.

Devin Desktop’s Cascade agent natively connects to user-configured MCP servers.

Cascade includes a dedicated web-search tool among its built-in agent capabilities.

Cascade supports browser-control actions with configurable auto-execution permissions.

Cascade can run terminal commands and install required packages as part of agent work.

Cascade edits code and executes terminal workflows inside the development environment.

Devin Desktop exposes memories and rules for reusable project context; exact persistence depends on configuration.

Cascade connects to third-party systems through MCP servers, with additional enterprise controls.

Devin Desktop is an installed development environment for major desktop operating systems.

Cascade operates directly inside Devin Desktop’s code-editor environment.

Replit Agent connects to Model Context Protocol (MCP) servers as an MCP client on Starter, Core, Teams, and Enterprise through pre-listed and custom server connections.

Replit Agent includes live web search and content fetching on Starter, Core, Teams, and Enterprise so apps can use current documentation and real-world data.

Replit Agent can test a built application automatically in a real browser when App Testing is enabled.

Replit Agent can run build and debugging workflows inside Replit’s hosted development environment.

Replit Agent writes, runs, debugs, and improves applications in Replit’s hosted runtime.

Replit Agent accepts user file uploads on Starter, Core, Teams, and Enterprise through the Conversation prompt box, including drag-and-drop and the plus-menu Upload a file action.

Replit Agent accepts image and screenshot inputs on Starter, Core, Teams, and Enterprise so users can attach visual references or ask it to refine a design from a screenshot.

Replit checkpoints preserve Agent conversation context, project state, environment configuration, and Agent memory.

Replit Agent supports extended autonomous runs, while availability and duration depend on selected mode and account limits.

Replit Agent can configure application integrations, but this review did not verify general MCP-client support.

Replit Agent runs primarily in Replit’s web and mobile cloud surfaces rather than a verified native desktop application.