Supported file types
You can attach files to any AI Chat message to give the assistant more context. The attachment types differ by how they are processed:
Images (JPEG, PNG, WebP, GIF, and other common image formats), sent directly to the AI model as visual context alongside your message. Maximum 10 MB per file.
PDFs, sent directly to the AI model as document context. Maximum 10 MB per file.
Audio and video files, routed through Speak AI's standard transcription pipeline. The file is uploaded as a new media item, transcribed, and the transcript is added as context for your chat. This uses your transcription quota exactly as a normal upload would.
How to attach a file
Click the paperclip icon in the chat input toolbar.
Select one or more files from your device. Images and PDFs can be attached in batches. Only one audio or video file can be attached per message.
Your attachments appear as chips above the input box. Remove any you do not want by clicking the X on its chip.
Type your message and send.
You can also paste an image directly from your clipboard into the chat input box instead of picking a file.
Transcription quota note
Audio and video attachments count against your transcription minutes just like any other upload. If you are on a Free Trial or Pay As You Go plan, charges apply at your plan's per-minute rate. Check your usage at any time under Account.
Related articles
The fastest way to reach us is the live chat in the app (the chat bubble in the bottom corner). You can also email [email protected].