Tabnine

FAQs

• Q. What is Tabnine?
  Tabnine is an AI-powered code assistant that helps developers write code faster and more accurately. it integrates into a variety of IDEs (e.g. VS Code, JetBrains, Eclipse, etc.) and provides features such as code autocompletion, test generation, refactoring, and documentation in real-time. it is characterized by a strong emphasis on security and privacy, maintaining the principles of localized execution, offline support, and no code storage, while providing powerful AI capabilities.
• Q. Is Tabnine free?
  some features are available as free trials.
  - Dev Preview plan: 14 days free for eligible users.
  - Dev plan: This is a paid plan, but you get a 30-day free trial.
  otherwise, a paid subscription is required.
• Q. What plan do I need to use Tabnine as a team?
  We recommend the Enterprise plan.
  it comes with management features (user policies, SSO, code validation, etc.) to help multiple people collaborate,
  Integrations with Jira, GitHub, GitLab, Bitbucket, and more to increase team productivity.
  IP protection and the ability to run models in your own environment make it ideal for security-critical enterprise environments.
• Q. Does Tabnine store or share my code?
  no. Tabnine never stores your code or shares it with third parties - it is only processed temporarily to generate real-time suggestions and is discarded immediately, not stored on our servers. We also do not use your code to train our AI models, and your data is strictly protected from external leakage.
• Q. How is Tabnine deployed, and do I have a choice?
  Tabnine supports both cloud hosting and local on-premises deployments. users can install it independently in a SaaS-based environment or on their own infrastructure, depending on their needs, and it can also operate completely offline in a virtual private network (VPC) environment. the deployment method is entirely under your control and can be chosen based on your security policies.
• Q. What AI models does Tabnine use, and can I choose my own?
  yes. In addition to our own models, Tabnine supports a variety of large language models (LLMs), including OpenAI, Anthropic, and more, and you can choose your own model or easily switch between them. This gives you the flexibility to utilize the best AI for your specific project or usage.
• Q. Will Tabnine use my code to train AI models?
  absolutely not - Tabnine only uses licensed, publicly available code for training, and neither your code nor your company's source code is included in training the model. For Enterprise customers, you can create custom models based on your code, but these are also private and run in your own environment.
• Q. How does Tabnine ensure security and privacy?
  Tabnine adheres to the highest security standards:
  - End-to-End Encryption
  - TLS communication protection
  - Compliance with international data protection regulations (GDPR, etc.)
  - No code retention (stateless processing)
  - Continuous security audits and threat monitoring
  this provides peace of mind for organizations working with sensitive source code.
• Q. Can Tabnine integrate with external tools like Jira?
  answer. Tabnine integrates with Atlassian Jira to provide issue-based code generation and validation. AI automatically generates code based on Jira issues and verifies that it meets the requirements.
• Q. Does Tabnine also offer AI code review?
  provide. Tabnine's code review agents provide real-time feedback on pull requests or code in the IDE based on your team's coding policies, rules, and security standards. if there are any deviations, it even provides fix suggestions to improve code quality.
• Q. Does Tabnine have legal protections for enterprise customers?
  answer. Tabnine provides legal protection for Enterprise customers, including IP Indemnification. this means that we minimize your liability for copyright disputes that may arise while using Tabnine, and provide safeguards to reduce your legal risk.