Generate Enterprise Troubleshooting and FAQ
This page answers frequently asked questions about Generate Enterprise and provides fixes for the most common issues. It covers what Generate is, which AI models and file formats it supports, how data is kept secure, and how to resolve login, chat, knowledge base, and connector problems.
What AI models does Generate Enterprise support?
Section titled “What AI models does Generate Enterprise support?”Generate Enterprise supports multiple AI providers, including OpenAI, Anthropic, Mistral, Groq, Cohere, and AWS Bedrock. Each organization’s administrator chooses which models are available to its users, so the exact set of selectable models depends on the organization’s configuration.
Is data in Generate Enterprise secure, and is it used to train AI models?
Section titled “Is data in Generate Enterprise secure, and is it used to train AI models?”Yes, data stays secure and is not used to train AI models. Documents and conversations remain within the organization’s environment, and administrators control data access and retention policies. Conversations are visible only to the user unless explicitly shared, though administrators may have oversight access depending on organization policy.
What file formats and size limits does the Knowledge Base support?
Section titled “What file formats and size limits does the Knowledge Base support?”The Generate Knowledge Base supports PDF, DOCX, PPTX, XLSX, CSV, and TXT files. Individual files can be up to 200 MB, and there is no limit on the number of files. Generate does not support uploading standalone image files, but it can process images embedded within supported documents, such as tables, charts, and infographics in PDFs and Word files. Most documents are processed in seconds to a few minutes depending on size.
Why is the AI giving incorrect or irrelevant answers?
Section titled “Why is the AI giving incorrect or irrelevant answers?”Answer quality depends on the Knowledge Base and on how the question is phrased. To improve answers: confirm relevant documents are uploaded and up to date, remove outdated documents that could mislead retrieval, be more specific and add context to the question, and try rephrasing. If using Projects, confirm the document is uploaded to the correct project.
What should you do if the AI is not responding or is slow?
Section titled “What should you do if the AI is not responding or is slow?”If the AI is not responding, check the internet connection, refresh the page, start a new conversation to rule out a conversation-specific issue, and check whether the AI provider is experiencing downtime. If responses are slow, note that speed depends on the chosen model and the complexity and length of the question; persistent slowness may warrant an administrator reviewing the model configuration.
How do you fix login and access problems in Generate?
Section titled “How do you fix login and access problems in Generate?”For login trouble, verify the email and password, use “Forgot password” on the login screen to reset, and check with an administrator in case the account was deactivated or not yet set up. If a verification email does not arrive, check spam/junk, confirm the correct address, wait a few minutes, and use Resend. An “Access Denied” message means the feature is not enabled for the account — contact an administrator to request access.
How do you fix document upload and processing issues?
Section titled “How do you fix document upload and processing issues?”If a document will not upload, confirm the format is supported (PDF, DOCX, PPTX, XLSX, CSV, TXT), confirm the file is under 200 MB, try a different browser, and check the connection. If a document is stuck in “Processing” for more than about 30 minutes, delete and re-upload it. If the AI is not using an uploaded document, confirm processing finished (status “Ready”), ask clearly related questions, and confirm it was uploaded to the correct project.
How do you fix connector and integration issues?
Section titled “How do you fix connector and integration issues?”If a connector will not connect, sign in with an account that has the right permissions, check whether an administrator must approve the integration first, and try disconnecting and reconnecting. If data from a connected service is not appearing, allow a few minutes for the initial sync, confirm the correct folders or channels were selected during setup, and use Reconnect if the connection has expired.
Related
Section titled “Related”- Feature details: see
chat-superagent,knowledge-base,integrations, andconfiguration-adminin this GEO set. - Full customer documentation:
support/faqandsupport/troubleshootingon the Generate Enterprise Mintlify docs site.