This endpoint is used to enqueue new AI assistant jobs. Each job type has their own set of arguments, which are documented in the job_info field of the /ai/jobs/types endpoint, as well as in the endpoint description below.
The valid job types are: deposition_summary, simple_summary, ab_content_creation, redline_summary, content_spinner, email_creator.

Simple Summary
Odin simplifies information processing by creating concise summaries from lengthy texts, capturing main points and key ideas swiftly and efficiently.
Required arguments:

objective: the objective prompt, explaining the task for the summary to the AI. For example: "Extract all key data points and information into a detailed summary."
model_name: the desired model name to use for the job. Options are: gpt-4-0613, gpt-3.5-turbo, gpt-3.5-turbo-16k-0613. Note that gpt-4 models are currently limited to subscription users only.

Deposition Summary
Odin proficiently creates deposition summaries, analyzes transcripts to highlight key evidence and testimonies, saving you time and effort.
Required arguments:

type_of_law: the desired type of law to use for the deposition summary. Examples include, but are not limited to:
- Civil Litigation Law
- Criminal Defense Law
- Corporate Law
- Employment Law
- Environmental Law
- Estate Planning Law
- Family Law
- Immigration Law
- Intellectual Property Law
- Real Estate Law
- Personal Injury Law

AB Content Creation
Create multiple versions of content based on your documents with unique tones. Great for email campaigns!
Required arguments:

instructions: the instructions for the AI to follow. For example: "Write a blog post about the benefits of Odin."
tones: the tones to use for the content creation. Multiple can be selected, supplied as an array of strings. Examples include, but are not limited to:
- Formal and Professional
- Friendly and Approachable
- Inquisitive and Curious
- Enthusiastic and Exciting
- Empathetic and Supportive
- Casual and Informal
- Concise and Direct
- Inspirational and Motivational
- Authoritative and Confident
- Educational and Informative
temperature: the desired creativity level to use for the content creation. A float between 0 and 1, with 0 being the least creative and 1 being the most creative.
model_name: the desired model name to use for the job. Options are: gpt-4-0613, gpt-3.5-turbo, gpt-3.5-turbo-16k-0613. Note that gpt-4 models are currently limited to subscription users only.

Redline Summary
Generate a summary of differences between two documents using Odin.
Required arguments:

model_name: the desired model name to use for the job. Options are: gpt-4-0613. Note that gpt-4 models are currently limited to subscription users only.
ignore_instructions: the instructions for the model on what changes to ignore. Provided as an array of strings.
ignore_pdf_headers_and_footers: whether to ignore PDF headers and footers. Provided as a boolean.
credit_usage_limit: the maximum number of credits to use for the job. Provided as an integer.
diff_ai_overide_limit: the maximum number of differences to track in a given block. Provided as an integer.

Content Spinner
Have Odin generate new content based on your original text and instructions!
Required arguments:

input_text: the input text to use for the content spinner. For example: "Odin is a great tool for summarizing documents."
temperature: the desired creativity level to use for the content spinner. A float between 0 and 1, with 0 being the least creative and 1 being the most creative.
model_name: the desired model name to use for the job. Options are: gpt-4-0613, gpt-3.5-turbo, gpt-3.5-turbo-16k-0613. Note that gpt-4 models are currently limited to subscription users only.

Optional arguments:

additional_instructions: the additional instructions for the content spinner. Provided as an array of strings.

Email Creator
Have Odin generate an email based on your instructions, either as plain text or HTML!
Required arguments:

sender: the sender of the email.
content_instructions: the instructions for the content of the email. Provided as an array of strings.
formatting_example: the formatting example for the email. For example: "Please format the email as follows: Greetings, [sender]! I hope you are doing well. [content] Sincerely, [recipient]."
model_name: the desired model name to use for the job. Options are: gpt-4-0613, gpt-3.5-turbo, gpt-3.5-turbo-16k-0613. Note that gpt-4 models are currently limited to subscription users only.

Optional arguments:

recipient: the recipient of the email.
generate_html: whether to generate the email as HTML. Provided as a boolean. False by default.

200Successful Response

422Validation Error