Jules
by Google
Starting at
Free tier (15 daily tasks, 3 concurrent); paid via Google AI Pro ($19.99/mo) and Ultra ($124.99/mo) for 5x/20x higher limits
Google's async AI coding agent built on Gemini 2.5 Pro.
Last verified: June 2026
What Jules does
Jules is Google's bet on asynchronous AI coding agents. Unlike Cursor or Copilot, which work inline while you type, Jules takes a task, disappears, and comes back with a PR. You assign it a GitHub issue, it clones the repo into a sandboxed Google Cloud environment, writes the code, runs the tests, and opens a pull request for your review.
It went out of beta on August 7, 2025, and has since handled over 140,000 code improvements for public projects during the beta period alone.
How it works
Jules reads from GitHub Issues, understands the codebase using Gemini 2.5 Pro's long-context window, and operates entirely in the background. You don't need to stay in a chat window. Assign a task, do other work, review the result when it's ready.
This makes it useful for a different workflow than most AI coding tools: not inline suggestions, but delegated tasks. Bug fixes, documentation updates, test coverage, dependency upgrades — work you'd otherwise add to a backlog.
Who uses it
Teams that want to offload predictable, well-scoped coding tasks. Solo developers handling backlog debt. Open-source maintainers who want help triaging and fixing reported issues. It's less suited for exploratory or architectural work where you need the back-and-forth of a conversation.
Pricing
Free tier available for individual use. Paid plans use a credit model billed per task, making it cost-predictable for teams with consistent workloads.
Pros
- + Runs on Gemini 2.5 Pro with strong reasoning for complex multi-file tasks
- + Works asynchronously so you can assign tasks and return to results
- + Native GitHub integration — reads issues and opens PRs directly
- + Sandboxed Google Cloud environment for safe code execution
- + Multimodal output with test visualization support
Cons
- - Still maturing — autonomous agent errors require manual review
- - Limited to GitHub (no GitLab or Bitbucket support at launch)
- - Per-task credit model can get expensive for heavy usage
- - Less control over execution steps than interactive tools like Cursor
What Users Actually Complain About
Autonomous execution means Jules can make incorrect assumptions about intent — always review the PR before merging. Currently GitHub-only. Paid plans are tied to Google AI subscriptions and, as of 2026, only available on individual @gmail.com accounts; Workspace/org upgrade paths are still rolling out.
Skip it if:
You need real-time collaborative coding or your repos are on GitLab or Bitbucket. Jules works best for well-defined, isolated tasks — not open-ended architectural work.
Based on community feedback from Reddit, HN, and G2 reviews.
Compare Jules with
Frequently Asked Questions
What is Jules?
Google's async AI coding agent built on Gemini 2.5 Pro.
How much does Jules cost?
Jules uses a freemium pricing model with plans starting at Free tier (15 daily tasks, 3 concurrent); paid via Google AI Pro ($19.99/mo) and Ultra ($124.99/mo) for 5x/20x higher limits.
What are the main advantages of Jules?
The key advantages of Jules include: Runs on Gemini 2.5 Pro with strong reasoning for complex multi-file tasks; Works asynchronously so you can assign tasks and return to results; Native GitHub integration — reads issues and opens PRs directly; Sandboxed Google Cloud environment for safe code execution; Multimodal output with test visualization support.
What are the drawbacks of Jules?
Some limitations to consider: Still maturing — autonomous agent errors require manual review; Limited to GitHub (no GitLab or Bitbucket support at launch); Per-task credit model can get expensive for heavy usage; Less control over execution steps than interactive tools like Cursor.
What category does Jules belong to?
Jules is a Code Assistant tool developed by Google.
Jules Comparisons
Jules vs GitHub Copilot
Jules (Google) vs GitHub Copilot — async AI coding agent vs inline AI assistant. Two very different approaches to AI-assisted development.
Jules vs Cursor
Jules (Google) vs Cursor — Google's async AI agent vs the leading AI-native IDE. Different tools for different workflows.
Code Assistant Guides
Best AI Code Assistants 2026
Best ToolsCompare the best AI code assistants for 2026. Expert analysis of GitHub Copilot, Cursor, Amazon Q Developer, and 10 more tools to boost developer productivity.
How to Choose an AI Code Assistant
How to ChooseExpert guide to selecting AI code assistants for DevOps teams. Compare GitHub Copilot, Cursor, Amazon Q Developer & more. Features, pricing & integration tips.
GitHub Copilot vs Cursor vs Windsurf: Which AI Code Editor Should You Use
Best ToolsCompare GitHub Copilot, Cursor, and Windsurf AI code editors. In-depth analysis of features, pricing, and performance to help you choose the best AI coding tool.
Best Free AI Code Assistants in 2026: Codeium, Supermaven, Continue & More
Best ToolsYou don't need to pay for GitHub Copilot. A new generation of free and open-source AI code assistants delivers serious productivity gains at zero cost — here's how to choose the right one.
Try Jules
Starting at Free tier (15 daily tasks, 3 concurrent); paid via Google AI Pro ($19.99/mo) and Ultra ($124.99/mo) for 5x/20x higher limits
Other Code Assistant Tools
View all 48 tools →Aider
Paul Gauthier
Aider is an AI-powered pair programming tool that works directly in your terminal to help you edit code in your existing git repositories.
Amazon Q Developer
Amazon Web Services
AI code companion from AWS with security scanning
Amp
Sourcegraph
Agentic coding tool from Sourcegraph (VS Code extension + CLI) that replaced the now-retired Cody, with a more autonomous approach to multi-file tasks.
Amplication
Amplication
Formerly Amplication (open-source backend code generator), the product rebranded to Overcut and pivoted to an agentic SDLC orchestration platform for deploying and coordinating AI agents across the software development lifecycle.