Messages represent the content that is sent to / returned from LLMs. They contain the message text, any associated files with the content, the author of the content, and the role in which the content applies.
Annotations at the Message level directly apply to this specific message. For example, the annotation for “response formatting” will be at this level.
The role of the sender in the conversation (e.g., user, assistant).
Available options:
system
,
user
,
assistant
,
function
The content of the message, including text and any attachments.
The textual content of the message.
The reference texts associated with the message.
content.reference_texts. content
Content of the reference text.
content.reference_texts. category
Category of the reference text.
content.reference_texts. url
URL source of the reference text.
Any files or attachments included with the message.
File Detailed Image File Detailed Audio File
content.attachments. content
The base64-encoded content of the file.
content.attachments. mime_type
The MIME type of the file, indicating its format.
content.attachments. scale_url
A URL pointing to the location of the file in Scale's storage.
A URL pointing to the source location of the file.
The ID of the source system or user that sent the message.
Integer Text Category Category-Multiple
Represents a generic annotation.
Unique identifier for the annotation.
The type of the value and the possible_values, if they exist.
Further details about the question.
Example: "Does the response contain issues with formatting?"
String representation of the possible options.
A string representation of the annotation.
Example: ["Major Issues", "Minor Issues", "No Issues"]
annotations.metadata. criteria
Option 1 · string Integer · object Text · object Category · object Category-Multiple · object
Unique identifier for an annotation.
Integer type annotation value.
annotations. possible_values
The possible values for this annotation.
Integer type annotation value.
The name of the model that generated the message.
model_parameters. temperature
The temperature of the model that generated the message.
model_parameters. max_completion_tokens
The maximum number of tokens the model can generate.