Google-backed coding tool Jules is now officially out of beta, marking its full availability to developers and teams worldwide. The move comes just over two months after its public preview launch at Google I/O 2025 in May 2025

What Is Jules and Why It Matters

Powered by Gemini 2.5 Pro, Google’s most advanced reasoning model tailored for coding tasks, Jules functions as a fully agentic, asynchronous coding assistant—not a simple autocomplete tool. It integrates with GitHub, clones entire repositories into secure Google Cloud VMs, and autonomously executes complex tasks such as bug fixes, feature implementation, dependency upgrades, and test writing—all while you continue working on other tasks

Today, Jules grows up. We're officially moving out of beta. This is a huge milestone, made possible by our incredible community of beta users. To celebrate, we're launching our most significant update yet. pic.twitter.com/7BCsXUf23f

— Jules (@julesagent) August 6, 2025

During its beta period, thousands of developers tackled tens of thousands of tasks with Jules, resulting in over 140,000 public code improvements submitted. Feedback from this period led to UI enhancements, bug fixes, GitHub Issues integration, multimodal support, and optimized task execution via previous environment snapshots

New Pricing Tiers and Usage Limits

With this release, Jules introduces structured access tiers:

• Free “introductory access” tier: Up to 15 tasks per day and 3 concurrent tasks (down from 60/day during beta).
• Google AI Pro subscribers: ~5× higher usage capacity.
• Google AI Ultra subscribers: Up to 20× more tasks per day for heavy workflows

These tiers reflect real-world usage insights collected over the beta phase, helping Google design fair access plans tailored to different user needs

Enhanced Developer Experience

Highlights of updated features include:

• Environment Snapshots to allow reused combinations of dependencies and install scripts for faster execution of recurring tasks
• GitHub Issues integration, enabling coding agent workflows that map directly to tracked issues.
• Multimodal support, allowing Jules to provide visual output previews or audio summaries of code changes.
• Audio changelogs, so developers can listen to updates rather than parse diffs manually

Jules also supports mobile web access, and Google is actively exploring feature enhancements for mobile platforms following feedback that nearly half of beta usage occurred on mobile browsing

Why Jules Stands Out

Unlike synchronous code‑completion tools like Cursor, Windsurf, or Copilot, Jules runs asynchronously in the cloud—meaning you can dispatch tasks and step away while it works inside a secure VM. You receive notifications or updates later, freeing you from needing to monitor the process in real time

Its agentic model offers deeper reasoning across codebases—understanding project context and structuring multi‑file changes intelligently. Each task is preceded by a visible plan and reasoning before execution. This level of transparency paired with autonomy marks a major leap toward agentic development workflows

Getting Started with Jules

Developers can start using Jules right away:

1. Sign in with a Google account.
2. Visit jules.google to connect GitHub repositories.
3. Select a repo and branch, and enter a task prompt such as “Fix login bug” or “Write unit tests for utils.js”.
4. Review the plan, approve or modify it, and let Jules run it asynchronously in a cloud VM.
5. Receive a pull request with diff and an audio or visual summary when it completes

Initially free during public beta, Jules will eventually adopt usage-based pricing aligned with the Pro and Ultra tiers of Google AI subscriptions.