model_catalog
Creates, updates, deletes, gets or lists a model_catalog resource.
Overview
| Name | model_catalog |
| Type | Resource |
| Id | digitalocean.genai.model_catalog |
Fields
The following fields are returned by SELECT queries:
- genai_get_model_catalog_card
- genai_list_model_catalog
A successful response.
| Name | Datatype | Description |
|---|---|---|
id | string | Identity (same as Entry) (example: "506a3371-88d0-4047-9212-e2079497ac68") |
name | string | (example: "Llama 3.1 70B") |
hugging_face_id | string | The Hugging Face repository ID (e.g. "meta-llama/Llama-3.3-70B-Instruct") the model is based on, when applicable. Omitted for models not sourced from Hugging Face. (example: "meta-llama/Llama-3.3-70B-Instruct") |
model_id | string | Model identifier used for API calls (e.g., "llama3.1-70b-instruct") (example: "llama3.1-70b-instruct") |
availability | array | |
badges | array | Badges for models |
benchmark_score | string | Benchmark scores for this model, stored as arbitrary JSON (opaque JSON object) |
capabilities | array | |
code_snippets | object | Code examples for using the model |
context_window | string (int64) | Specs (same as Entry) (example: 128000) |
creator | string | Model creator/developer (e.g., "Meta", "Anthropic", "OpenAI") (example: "Meta") |
description | string | Card-specific (example: "Claude Sonnet 4 is Anthropic's most capable model...") |
max_output_tokens | string (int64) | The maximum number of output tokens the model can generate in a single response. (example: 65536) |
modalities | object | Input/output modalities |
parameter_count | number (float) | |
pricing | object | Pricing per million tokens (aligns with existing ModelPrice pattern) |
pricing_detail | object | The complete set of prices for a model, covering every available variant. |
provider | string | (MODEL_PROVIDER_DIGITALOCEAN, MODEL_PROVIDER_ANTHROPIC, MODEL_PROVIDER_OPENAI) (default: MODEL_PROVIDER_DIGITALOCEAN, example: MODEL_PROVIDER_DIGITALOCEAN) |
short_description | string | (example: "Fast, efficient model for general tasks") |
type | string | (example: "text-to-text") |
A successful response.
| Name | Datatype | Description |
|---|---|---|
id | string | Identity (example: "506a3371-88d0-4047-9212-e2079497ac68") |
name | string | (example: "Llama 3.1 70B") |
hugging_face_id | string | The Hugging Face repository ID (e.g. "meta-llama/Llama-3.3-70B-Instruct") the model is based on, when applicable. Omitted for models not sourced from Hugging Face. (example: "meta-llama/Llama-3.3-70B-Instruct") |
model_id | string | Model identifier used for API calls (e.g., "llama3.1-70b-instruct") (example: "llama3.1-70b-instruct") |
availability | array | |
badges | array | Badges for models |
benchmark_score | string | Benchmark scores for this model, stored as arbitrary JSON (opaque JSON object) |
capabilities | array | |
context_window | string (int64) | Specs (flat) (example: 128000) |
created_at | string (date-time) | RFC 3339 timestamp indicating when the model was added to the catalog. (example: 2021-01-01T00:00:00Z) |
creator | string | Model creator/developer (e.g., "Meta", "Anthropic", "OpenAI") (example: "Meta") |
max_output_tokens | string (int64) | The maximum number of output tokens the model can generate in a single response. (example: 65536) |
parameter_count | number (float) | |
pricing | object | Pricing per million tokens (aligns with existing ModelPrice pattern) |
provider | string | (MODEL_PROVIDER_DIGITALOCEAN, MODEL_PROVIDER_ANTHROPIC, MODEL_PROVIDER_OPENAI) (default: MODEL_PROVIDER_DIGITALOCEAN, example: MODEL_PROVIDER_DIGITALOCEAN) |
scaled_pricing_enabled | boolean | True when this model's pricing varies over time. Retrieve the model's details for the full pricing schedule. False for models with fixed pricing. |
short_description | string | (example: "Fast, efficient model for general tasks") |
type | string | (example: "text-to-text") |
Methods
The following methods are available for this resource:
| Name | Accessible by | Required Params | Optional Params | Description |
|---|---|---|---|---|
genai_get_model_catalog_card | select | id | model_id | Returns detailed information for a specific model in the catalog including capabilities, pricing, and code examples. |
genai_list_model_catalog | select | page, limit | Returns all available models. |
Parameters
Parameters can be passed in the WHERE clause of a query. Check the Methods section to see which parameters are required or optional for each operation.
| Name | Datatype | Description |
|---|---|---|
id | string | (example: "example string") |
limit | integer | (example: 1) |
model_id | string | Model identifier used for API calls (e.g., "llama3.1-70b-instruct"). Alternative to UUID lookup. (example: "example string") |
page | integer | (example: 1) |
SELECT examples
- genai_get_model_catalog_card
- genai_list_model_catalog
Returns detailed information for a specific model in the catalog including capabilities, pricing, and code examples.
SELECT
id,
name,
hugging_face_id,
model_id,
availability,
badges,
benchmark_score,
capabilities,
code_snippets,
context_window,
creator,
description,
max_output_tokens,
modalities,
parameter_count,
pricing,
pricing_detail,
provider,
short_description,
type
FROM digitalocean.genai.model_catalog
WHERE id = '{{ id }}' -- required
AND model_id = '{{ model_id }}'
;
Returns all available models.
SELECT
id,
name,
hugging_face_id,
model_id,
availability,
badges,
benchmark_score,
capabilities,
context_window,
created_at,
creator,
max_output_tokens,
parameter_count,
pricing,
provider,
scaled_pricing_enabled,
short_description,
type
FROM digitalocean.genai.model_catalog
WHERE page = '{{ page }}'
AND limit = '{{ limit }}'
;