Google has officially launched its AI coding assistant Jules and taken it out of beta. The tool, which was introduced as a public preview at the end of May, can now be used as a regular product.
This was reported by TechCrunch and Golem. Jules runs on Gemini 2.5 Pro and is designed as an asynchronous, agent-based programming assistant that supports developers in a variety of tasks. The system integrates with GitHub, clones repositories to a virtual cloud environment within Google Cloud, and independently performs operations on the code. This allows developers to outsource multiple processes without having to be actively involved in every step.
Where traditional AI tools often work synchronously and require direct interaction, Jules distinguishes itself by performing tasks completely asynchronously. A developer can initiate tasks, shut down their laptop, and return later to retrieve the results. This functionality makes Jules particularly suitable for long-running or routine tasks that do not require continuous monitoring.
During the beta phase, Jules already supported thousands of developers. The platform was used for tasks such as detecting and resolving bugs, expanding existing projects, and performing maintenance work based on previously generated code. In practice, Jules proved particularly effective for repetitive development tasks, such as creating test scripts, adjusting dependencies, and performing structural refactorings. In addition to these applications, the agent also offers support in generating changelogs. If desired, the tool can also read these logs aloud, which increases its usability in contexts where visual attention is limited.
Limitations of the free version
The transition to regular rollout will be accompanied by a modified pricing structure. Users get access to a free entry-level version, which allows them to perform fifteen tasks per day, with a maximum of three simultaneous processes. This is a significant reduction from the sixty tasks per day that were available during the beta. For more intensive use, Pro and Ultra subscriptions are available, which are part of Google’s broader AI offering. In Europe, these are priced at €21.99 and €274.99 per month. They offer five and twenty times more processing capacity, respectively. This applies to both the number of tasks and the number of parallel sessions.
The pricing structure is based on data collected during the test period. With the original limit, Google wanted to gain insight into actual usage behavior in order to make targeted choices in product positioning. The current limits in the free version are intended to give developers a realistic picture of how Jules can be used in concrete projects.
The privacy terms and conditions have been tightened and clarified. Public repositories can be used to train the AI, while private projects are strictly kept out of that process. They are not analyzed or shared, but are processed via Google Cloud. The latter means that, technically speaking, every command runs through Google’s infrastructure, which may cause some developers to be cautious, despite explicit guarantees of data minimization.
Thousands of code improvements
During the beta, thousands of code improvements were generated based on tens of thousands of tasks. This phase provided valuable insights, such as the need to be able to reuse previous setups for faster execution, support for multimodal input, and deeper integration with GitHub, including the automatic opening of pull requests. The platform has also been modified to function better without an existing codebase, allowing users to get started even with an empty repository. This significantly increases accessibility, especially for users who are experimenting with AI without already having a project in place.
Usage statistics show that Jules has now registered millions of visits worldwide. A large proportion of these came from mobile devices. Despite the lack of a native mobile app, many developers use the platform via the web interface on their phones.
Internally, Google now also uses Jules for its own development projects. Google sees Jules as a strategic building block within its development stack.