Model
Snyk API reference for the Model endpoints, including request parameters and response schemas
List AI models in Snyk's Risk Database, with their metadata and risk scores.
Results are sorted by last_updated, most recent first.
Risk scores, capability scores, licensing, and country of origin are returned only if your tenant is entitled to Evo Risk Intelligence. Otherwise, these fields are empty or omitted.
Required permissions
View Tenant Details (tenant.read)
API key value must be prefixed with "Token ".
Tenant ID
e4a039e5-968b-4f1a-b190-caf3514243b5Requested API version
2024-10-15Pattern: ^(wip|work-in-progress|experimental|beta|((([0-9]{4})-([0-1][0-9]))-((3[01])|(0[1-9])|([12][0-9]))(~(wip|work-in-progress|experimental|beta))?))$Return only models whose provider slug matches one of the given values (case-insensitive). Comma-separated, e.g. provider=openai,anthropic.
["openai","anthropic"]When true, return only models that have risk scores; when false, only models without them.
trueWhen true, return only models whose risk assessment is complete; when false, only models whose assessment is missing or partial.
trueReturn only models whose last_updated is after this date and time. last_updated is when the model's data was last updated in Snyk's Risk Database.
2025-01-01T00:00:00ZReturn only the model with this name: its source ID, an alias, or a provider-prefixed name, for example, openai/gpt-4. Matching is case-insensitive.
meta-llama/Llama-3.1-8B-InstructReturn the page of results immediately after this cursor
v1.eyJpZCI6IjEwMDAifQo=Return the page of results immediately before this cursor
v1.eyJpZCI6IjExMDAifQo=Number of results to return per page
10Example: 10Include a count of the total matching items in the response meta
withPossible values: The models have been successfully retrieved.
Bad Request: A parameter provided as a part of the request was invalid.
Not Found: The resource being operated on could not be found or you do not have permission to access it.
Internal Server Error: An error was encountered while attempting to process the request.
GET /rest/tenants/{tenant_id}/risk_db/models?version=text HTTP/1.1
Host: api.snyk.io
Authorization: YOUR_API_KEY
Accept: */*
{
"data": [
{
"attributes": {
"access_info": {
"api_available": true,
"cost_tier": "medium",
"license_name": "Llama-3.1",
"license_type": "proprietary",
"self_hostable": true
},
"architectures": [
"LlamaForCausalLM"
],
"capability_assessment": {
"eci_score": 128.5
},
"context_length": 131072,
"country_of_origin": "US",
"has_risk_assessment": true,
"inputs": [
"text"
],
"last_updated": "2024-09-25T17:00:57Z",
"model_family": "llama-3-1",
"model_type": "llama",
"model_variant": "llama-3-1-8b-instruct",
"outputs": [
"text"
],
"parameters": "8B",
"pipeline_tag": "text-generation",
"provider": "meta",
"provider_name": "Meta Llama",
"release_date": "2024-07-18T08:56:00Z",
"risk_assessment": [
{
"attacker_goals": [
{
"name": "insecure_code_generation",
"score": 579
}
],
"name": "code_generation",
"score": 579
}
],
"risk_assessment_complete": true,
"source_id": "meta-llama/Llama-3.1-8B-Instruct",
"source_url": "https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct",
"tasks": [
"text-generation"
]
},
"id": "0b7a2c1e-5d3f-4e8a-9c6b-1f2e3d4c5b6a",
"type": "model"
}
],
"jsonapi": {
"version": "1.0"
},
"links": {
"first": "https://example.com/api/resource?ending_before=v1.eyJpZCI6IjExIn0K",
"last": "https://example.com/api/resource?starting_after=v1.eyJpZCI6IjMwIn0K",
"next": "https://example.com/api/resource?starting_after=v1.eyJpZCI6IjEwIn0K"
},
"meta": {
"count": 42,
"risk_taxonomy": {
"categories": [
{
"description": "Does it compromise code security?",
"display_name": "Code Security",
"name": "code_security",
"sub_categories": [
{
"attacker_goals": [
{
"description": "Generate code containing security vulnerabilities when asked for help with coding tasks",
"display_name": "Insecure code generation",
"name": "insecure_code_generation",
"taxonomy_mappings": {
"LLM": [
"LLM05"
]
}
}
],
"description": "Vulnerabilities introduced when the agent writes code",
"display_name": "Code Generation",
"name": "code_generation"
}
]
}
]
}
}
}Get a single AI model from Snyk's Risk Database by ID, with its metadata and risk scores.
To look a model up by name, list models with the model_name filter.
Risk scores, capability scores, licensing, and country of origin are returned only if your tenant is entitled to Evo Risk Intelligence. Otherwise, these fields are empty or omitted.
Required permissions
View Tenant Details (tenant.read)
API key value must be prefixed with "Token ".
Tenant ID
e4a039e5-968b-4f1a-b190-caf3514243b5The ID of the model in Snyk's Risk Database.
0b7a2c1e-5d3f-4e8a-9c6b-1f2e3d4c5b6aRequested API version
2024-10-15Pattern: ^(wip|work-in-progress|experimental|beta|((([0-9]{4})-([0-1][0-9]))-((3[01])|(0[1-9])|([12][0-9]))(~(wip|work-in-progress|experimental|beta))?))$The model has been successfully retrieved.
Bad Request: A parameter provided as a part of the request was invalid.
Not Found: The resource being operated on could not be found or you do not have permission to access it.
Internal Server Error: An error was encountered while attempting to process the request.
GET /rest/tenants/{tenant_id}/risk_db/models/{model_id}?version=text HTTP/1.1
Host: api.snyk.io
Authorization: YOUR_API_KEY
Accept: */*
{
"data": {
"attributes": {
"access_info": {
"api_available": true,
"cost_tier": "medium",
"license_name": "Llama-3.1",
"license_type": "proprietary",
"self_hostable": true
},
"architectures": [
"LlamaForCausalLM"
],
"capability_assessment": {
"eci_score": 128.5
},
"context_length": 131072,
"country_of_origin": "US",
"has_risk_assessment": true,
"inputs": [
"text"
],
"last_updated": "2024-09-25T17:00:57Z",
"model_family": "llama-3-1",
"model_type": "llama",
"model_variant": "llama-3-1-8b-instruct",
"outputs": [
"text"
],
"parameters": "8B",
"pipeline_tag": "text-generation",
"provider": "meta",
"provider_name": "Meta Llama",
"release_date": "2024-07-18T08:56:00Z",
"risk_assessment": [
{
"attacker_goals": [
{
"name": "insecure_code_generation",
"score": 579
}
],
"name": "code_generation",
"score": 579
}
],
"risk_assessment_complete": true,
"source_id": "meta-llama/Llama-3.1-8B-Instruct",
"source_url": "https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct",
"tasks": [
"text-generation"
]
},
"id": "0b7a2c1e-5d3f-4e8a-9c6b-1f2e3d4c5b6a",
"type": "model"
},
"jsonapi": {
"version": "1.0"
},
"links": {
"first": "https://example.com/api/resource",
"last": "https://example.com/api/resource",
"next": "https://example.com/api/resource",
"prev": "https://example.com/api/resource",
"related": "https://example.com/api/resource",
"self": "https://example.com/api/resource"
},
"meta": {
"risk_taxonomy": {
"categories": [
{
"description": "Does it compromise code security?",
"display_name": "Code Security",
"name": "code_security",
"sub_categories": [
{
"attacker_goals": [
{
"description": "Generate code containing security vulnerabilities when asked for help with coding tasks",
"display_name": "Insecure code generation",
"name": "insecure_code_generation",
"taxonomy_mappings": {
"LLM": [
"LLM05"
]
}
}
],
"description": "Vulnerabilities introduced when the agent writes code",
"display_name": "Code Generation",
"name": "code_generation"
}
]
}
]
}
}
}Last updated
Was this helpful?