File Attachment
Last updated
Last updated
The File Attachment feature allows users to upload files as part of their prompt inputs when interacting with the application. This feature is designed to enhance the prompting experience by enabling users to provide additional context, data, or reference materials that can be utilized by the model to generate more accurate and relevant responses.
The File Attachment feature allows users to upload files as part of their prompts, providing additional context or reference materials for more accurate responses. Unlike the Knowledge Base or Retrieval-Augmented Generation (RAG) methods, which send only relevant information, this feature sends the entire file content to the model, ensuring a comprehensive input for generating responses.
Open your Sandbox.
Click the + button to add your file or URLs to the prompt.
Please note that our file attachment feature currently supports the following formats: .png, .jpg, .jpeg, .webp, .gif, .pdf, .csv, and .txt. You can upload up to 5 files per prompt, with each file limited to 50 MB.
After attaching your file, enter your desired prompt. Then, click Run all to generate the model's response based on your prompt.
Please note that the uploaded file will be visible to all users within the same workspace and will remain accessible until deleted by the user.
