AI in JetBrains IDEs includes a range of capabilities that can be accessed through the AI Assistant plugin.
Key capabilities include:
AI-powered assistance for writing and modifying code as you type.
Work with multiple agents, including Junie by JetBrains and third-party agents such as Claude Agent, OpenAI Codex, and Gemini CLI. Or connect and use any ACP-compatible agent with your own subscription.
Explain code, suggest refactoring, detect issues, and generate documentation directly in the editor.
Get assistance in areas such as version control, runtime error analysis, and database-related tasks.
Availability of some features may vary depending on the IDE and its version. Refer to the documentation for details.
Access to AI features may be restricted in certain countries. You can check the full list of supported locations here. Some features may function differently or be unavailable in Mainland China due to regional compliance requirements.
JetBrains IDEs can use a mix of JetBrains models, third-party cloud models, and user-provided or local models.
JetBrains AI provides access to a curated set of cloud models from leading providers, including OpenAI, Google, Anthropic, and Grok. These models are automatically selected for specific features or can be chosen in AI chat.
Code completion and next edit suggestions features rely on JetBrains models like Mellum, optimized specifically for coding tasks.
You can also connect and use models outside of JetBrains AI via API key or OAuth (for Codex agent), or connect local models via tools like Ollama or LM Studio
For the current list of supported cloud and custom-model options, see the Supported LLMs documentation.
In Mainland China, AI features are powered exclusively by Alibaba's Tongyi Qianwen (通义千问) large language models.
JetBrains AI features are available in all JetBrains IDEs, as well as ReSharper and Android Studio. However, not all features and JetBrains AI subscription tiers are supported in every product. For example, local code completion, which is available out of the box in most JetBrains IDEs, is not supported in IntelliJ IDEA without an Ultimate subscription, PyCharm without a Pro subscription, ReSharper, or Android Studio. In addition, it’s currently not possible to choose LLM providers or connect local models in ReSharper.
AI Free is not available in Android Studio, ReSharper, IntelliJ IDEA without an Ultimate subscription, and PyCharm without a Pro subscription.
Click the JetBrains AI icon in the top right corner of your IDE, From there, you can enable AI Assistant and access supported features.
When you enable JetBrains AI for the first time, you receive a free 30-day trial of AI Pro. Alternatively, you can start using AI features without a JetBrains AI subscription:
Log in with your provider account (OAuth) or use your own API key (BYOK) to access supported models.
Use external agents (ACP)
Connect and work with ACP-compatible agents using your own subscription.
You can connect your own local LLMs (via Ollama, LM Studio, and other OpenAI-compatible servers) to enable local AI chat*.You can enable them in the IDE settings (Settings | AI Assistant | Models).
*Some features, like next edit suggestions and Junie do not support local models.
Yes. JetBrains IDEs support Bring Your Own Key (BYOK), which lets you connect your own API keys from supported AI providers and use their models directly inside JetBrains IDEs.
You can use API keys from OpenAI, Anthropic, and other OpenAI-compatible providers in AI Chat, Junie, and Claude agents. No JetBrains AI subscription is required to use BYOK for chat and agents. However, to access other AI features, such as cloud code completion, you need to sign in to JetBrains AI.
You can easily manage how AI is used in your IDE, as all JetBrains IDEs require you to opt in before they provide access to any cloud-based AI functionality. If you don’t want cloud-based AI support, you can enable offline mode or disable AI features completely. See the documentation for more details.
Access to JetBrains AI Assistant and Junie is currently restricted in certain countries due to the limitations imposed by third-party AI providers (such as OpenAI, Anthropic, and Google). You can check the full list of supported locations here.
Yes, JetBrains AI is available in China, and it works exclusively with Alibaba's Tongyi Qianwen (通义千问) large language models. This integration ensures compliance with local regulations and offers optimized performance for Chinese users.
AI Assistant is exclusively available through the AI Pro subscription plan in China and offers code completion, code generation, AI-driven explanations, and more. Please note that certain AI features may function differently or be unavailable in China due to regional compliance requirements. Specifically, Junie, the JetBrains coding agent, is not available in China at this time.
We encourage you to try JetBrains AI features and share your feedback to help us improve. To leave feedback about AI Assistant, you can:
If you have questions or feedback about Junie, feel free to email us at junie@jetbrains.com.
We also encourage you to leave a review on JetBrains Marketplace for Junie or AI Assistant.
JetBrains AI offers AI Free*, AI Pro, AI Ultimate, and AI Enterprise. The main difference between the plans is the amount of AI Credits included for features powered by third-party AI providers.
See the pricing page for plan details and usage quotas.
* AI Free is not available in Android Studio, ReSharper and Intellij IDEA without Ultimate subscription and PyCharm without Pro subscription.
In Mainland China, JetBrains AI Assistant is accessible only through the AI Pro subscription plan.
Yes. A 30-day AI Pro trial is available when you first activate JetBrains AI. After the trial ends, you can either continue on the AI Free plan (if available in your IDE*) or upgrade to AI Pro or AI Ultimate for more usage.
* AI Free is not available in Android Studio, ReSharper and Intellij IDEA without Ultimate subscription and PyCharm without Pro subscription.
In Mainland China, JetBrains AI Assistant is accessible only through the AI Pro subscription plan.
AI Free is available to all users with active JetBrains IDE licenses, including educational licenses, classroom licenses, open source licenses, IDE trial licenses, licenses for users in the Startup Program, and other complimentary licenses.
As part of the AI Free plan, you get:
However, the AI Free plan is not available in IntelliJ IDEA without an Ultimate subscription, PyCharm without a Pro subscription, ReSharper, or Android Studio. To use AI in those products, you must have an AI Pro subscription.
AI Pro is included at no additional cost for:
*Al Pro is not included in complimentary All Products Pack subscriptions for students, teachers, and classrooms, or for Open Source program participants. In China, the AI Pro subscription is not included with the All Products Pack.
** If you get AI Pro with DataSpell, it is tied to that product only and cannot be used with other JetBrains products.
JetBrains AI subscription tiers include a fixed monthly quota of AI Credits, which regulate access to AI agents and features powered by third-party AI models.
Code completion and next edit suggestions are powered by JetBrains’ tailored models, and therefore do not consume AI quota.
The AI quota matches its price, meaning 1 AI Credit is equivalent to USD 1.00 (charged in your local currency). You get AI Credits equal to the value of your plan, plus a small bonus with AI Ultimate. For example, an individual AI Ultimate subscription costs USD 30.00 per month and now comes with USD 35.00 in AI Credits (USD 5 bonus credits).
Yes. You can track your remaining AI quota and any top-up AI Credits in the JetBrains AI widget in the IDE. Consumption varies depending on the models and features you use, and charged at LLM rates (in USD).
When your monthly AI quota is exhausted, cloud-powered JetBrains AI features pause until your quota renews. ou can buy additional AI Credits to continue using AIfeatures. Your included quota resets every 30 days from the date you first activated your JetBrains AI license, which may not match your billing cycle.
Top-up AI Credits are additional credits you can purchase to extend your cloud AI usage beyond the quota included with your JetBrains AI subscription. They require an active JetBrains AI license and:
- remain valid for 12 months
- for organizations they are shared across the team’s JetBrains AI licenses.
No. Any unused included quota does not roll over. The AI quota resets every 30 days from the date you first activated your JetBrains AI license. However, if you purchased Top-up AI Credits, they remain valid for 12 months.
Yes. Once you purchase a JetBrains AI subscription, it is linked to your JetBrains Account and can be used across all supported IDEs. Because your AI subscription is linked to your JetBrains Account, your AI quota will be shared across all products.
Yes. With Bring Your Own Key (BYOK), you can connect your own API keys from supported AI providers and use their models directly inside JetBrains IDEs.
When you use BYOK, billing is handled entirely by the AI provider whose API key you connect. JetBrains does not charge for BYOK usage, and all costs depend on the provider’s pricing and your actual model usage.
However, BYOK does not fully replace JetBrains AI. Some IDE-integrated features, such as cloud code completion and other advanced editor features, require signing in to JetBrains AI. You can use BYOK and JetBrains AI together to get full feature coverage.
Data sharing is opt-in for all paid licenses, including individual and company licenses (this also includes EAP and free trials). For company licenses, this setting is centrally maintained by your admins and cannot be changed individually.
For licenses that are for non-commercial use, this setting is enabled by default, but you can opt out at any time in the settings: Appearance & Behavior -> System Settings -> Data Sharing -> Allow detailed data collection by JetBrains AI.
For detailed information about our approach to data collection, please refer to our Data Collection and Use Policy.
JetBrains can collect two categories of data: anonymous telemetry about feature usage, and – if detailed data collection is enabled—more detailed code-related data such as prompts, responses, code snippets, edit history, terminal usage, and interactions with AI features. Detailed code-related data is used for product improvement and training JetBrains models, and it is not shared with third parties.
For detailed information about our approach to data collection, please refer to our Data Collection and Use Policy.
Only if you explicitly allow it. If you choose to share data with us, this data is used exclusively to improve JetBrains tools and train our own models, like Mellum, and it is never shared with third parties.
JetBrains does not persist any customer data unless a user opts-in to Detailed Data Collection. If you do, the shared data is retained for up to one year. Any data you share will be handled responsibly, in full compliance with all EU data protection laws.
When you use AI-powered features, the IDE sends your prompts to the LLM provider. In addition to your prompts, the IDE may add information about the context, such as pieces of code, file types, and frameworks used. This allows the LLM to provide the most relevant suggestions. JetBrains does not work with the large language model providers that use customer data for training models.
JetBrains AI uses AWS and Google Cloud as a core services layers. As for location, the AWS servers and Google Cloud servers are currently located in the EU, the US, and Asia.
You can use AI Assistant in offline mode, which allows you to connect locally hosted large language models (LLMs) via LM Studio or Ollama.
For organizations requiring deeper customization and control, AI Enterprise offers on-premises installation as part of JetBrains IDE Services, as well as the ability to connect on-premises models via Hugging Face, enabling a secure air-gapped environment. This ensures full control over data and model management, allowing companies to choose specific LLM providers.
You can disable AI assistance completely or control AI access on a project or file level. Use an .aiignore file to prevent specific files or folders from being sent to the AI. This gives you more granular control over privacy-sensitive code or repositories. You can learn more in the documentation.