API Documentation
FlashDocs Document Configuration
- GETList Document Configurations
- POSTCreate Google Slides FlashDocs Template/Library
- POSTCreate PowerPoint FlashDocs Template/Library
- GETGet Document Configuration Full Data
- PUTUpdate Document Configuration
- DELDelete Document Configuration
- GETGet Slide Configuration Metadata
- POSTCompute Markdown Mapping for Slide
- POSTValidate Text Placeholder Manual Insertion
Generate
Generate Slide(s)
The Generate Deck endpoint creates a slide deck based on user-defined parameters that specify its content and structure. Upon successful creation, the endpoint returns a link to the deck along with execution details. Unlike the POST /v3/generate/deck/task
endpoint—which immediately returns a deck generation task for polling—this endpoint waits until the entire deck has been rendered before sending the response.
POST
/
v3
/
generate
/
deck
Copy
curl --request POST \
--url https://api.flashdocs.ai/v3/generate/deck \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '{
"prompt": "In a few slides, describe the recent Innovations in Renewable Energy. Discuss the foundational problems and key papers. Include specific examples.",
"source_document_id": "abc-def-ghi-jkl",
"number_slides": 5,
"audience_domain": "slack.com",
"presentation_name": "Renewable Energy Innovations 2025",
"outline": [
{
"content_instruction": "Introduce the topic of renewable energy and its importance.",
"layout_instruction": "Title slide with a background image.",
"markdown": "# Parallel Inference on GPUs for LLMs\n## **Data Parallelism**\n- Distributes batches of input data across GPUs.\n- **Benefits:** Increased throughput via simultaneous processing.\n## **Model Parallelism**\n- Splits the model'\''s layers or components among GPUs.\n## **Pipeline Parallelism**\n- Overlaps different stages of inference, reducing idle time.\n- **Advantage:** Maximizes resource utilization.\n - Use various layers of DNN see [here](https://developer.nvidia.com/blog).",
"text_placeholder_manual_insertions": [
{
"placeholder": "[title]",
"value": "Renewable Energy Innovations 2027"
},
{
"placeholder": "[content]",
"value": "Introduce the topic of renewable energy and its importance."
}
],
"image_placeholder_manual_insertions": [
{
"alt_text": "[top right]",
"slide_index": 0,
"url": "https://example.com/image.jpg"
},
{
"alt_text": "[bottom left]",
"slide_index": 1,
"url": "https://example.com/image2.jpg"
}
],
"slide_id": "slide_123"
}
],
"export_options": {
"google_slides_permission_options": {
"anyone_edit_with_link": false,
"anyone_view_with_link": false,
"google_document_editors": [
"bob@acme.com",
"joe@acme.com"
]
},
"output_format": "google_slides_direct_link"
}
}'
Copy
{
"task_id": "task_67890",
"link_to_deck": "https://docs.google.com/presentation/d/abcdefg123456",
"runtime": 4.3778,
"success": true,
"error": ""
}
Authorizations
The access token received from the authorization server in the OAuth 2.0 flow.
Body
application/json
Response
200
application/json
Successful Response
The response is of type object
.
Copy
curl --request POST \
--url https://api.flashdocs.ai/v3/generate/deck \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '{
"prompt": "In a few slides, describe the recent Innovations in Renewable Energy. Discuss the foundational problems and key papers. Include specific examples.",
"source_document_id": "abc-def-ghi-jkl",
"number_slides": 5,
"audience_domain": "slack.com",
"presentation_name": "Renewable Energy Innovations 2025",
"outline": [
{
"content_instruction": "Introduce the topic of renewable energy and its importance.",
"layout_instruction": "Title slide with a background image.",
"markdown": "# Parallel Inference on GPUs for LLMs\n## **Data Parallelism**\n- Distributes batches of input data across GPUs.\n- **Benefits:** Increased throughput via simultaneous processing.\n## **Model Parallelism**\n- Splits the model'\''s layers or components among GPUs.\n## **Pipeline Parallelism**\n- Overlaps different stages of inference, reducing idle time.\n- **Advantage:** Maximizes resource utilization.\n - Use various layers of DNN see [here](https://developer.nvidia.com/blog).",
"text_placeholder_manual_insertions": [
{
"placeholder": "[title]",
"value": "Renewable Energy Innovations 2027"
},
{
"placeholder": "[content]",
"value": "Introduce the topic of renewable energy and its importance."
}
],
"image_placeholder_manual_insertions": [
{
"alt_text": "[top right]",
"slide_index": 0,
"url": "https://example.com/image.jpg"
},
{
"alt_text": "[bottom left]",
"slide_index": 1,
"url": "https://example.com/image2.jpg"
}
],
"slide_id": "slide_123"
}
],
"export_options": {
"google_slides_permission_options": {
"anyone_edit_with_link": false,
"anyone_view_with_link": false,
"google_document_editors": [
"bob@acme.com",
"joe@acme.com"
]
},
"output_format": "google_slides_direct_link"
}
}'
Copy
{
"task_id": "task_67890",
"link_to_deck": "https://docs.google.com/presentation/d/abcdefg123456",
"runtime": 4.3778,
"success": true,
"error": ""
}
Assistant
Responses are generated using AI and may contain mistakes.