Develop

API reference

Every endpoint, every parameter, and every response shape — typed and versioned.

Base URL

text

https://api.nodedata.dev/v1

All endpoints accept JSON bodies and return JSON responses. Errors follow the same shape across the API.

Versioning

Breaking changes ship under a new path prefix (/v1 → /v2). Additive changes (new fields, new optional parameters) ship in place without bumping the version. Removed or renamed fields are announced in the changelog at least 90 days before removal.

Error format

json

{
  "error": {
    "code": "validation_failed",
    "message": "license must be one of MIT, Apache-2.0, CC-BY-4.0",
    "field": "license",
    "request_id": "req_01HZ..."
  }
}

Always log request_id. Support requests resolve much faster when the request ID is included.

Pagination

List endpoints return a cursor in next_cursor when more results exist. Pass it back as ?cursor= on the next call.

json

{
  "items": [/* ... */],
  "next_cursor": "eyJsYXN0X2lkIjoibWRsXzAxSF...",
  "has_more": true
}

Rate limits

Surface	Limit
Read endpoints	1,000 req / minute / key
Write endpoints	120 req / minute / key
Upload endpoints	30 req / minute / key (uploads themselves are unmetered)
Inference (hosted)	Per listing — see the listing's gateway docs

Every response carries X-RateLimit-Remaining and X-RateLimit-Reset headers. When throttled the API responds with 429 and a Retry-After header in seconds.

Account

Get the authenticated account

http

GET /v1/me
Authorization: Bearer <key>

Response — 200
{
  "id": "usr_01HZ...",
  "slug": "acme",
  "name": "ACME Robotics",
  "email": "team@acme.dev",
  "created_at": "2026-03-19T18:04:11Z",
  "key": { "mode": "test", "scopes": ["models:read"] }
}

Smallest possible call to verify a key. Returns the account that owns the key plus the key's mode and scopes.

Models

List models

http

GET /v1/models
Authorization: Bearer <key>

Query parameters
  q          string   Full-text search query
  task       enum     manipulation | navigation | perception | language | vla
  framework  enum     pytorch | tensorflow | jax | onnx | gguf
  license    enum     MIT | Apache-2.0 | CC-BY-4.0 | Commercial
  cursor     string   Pagination cursor returned in the previous response
  limit      int      1–100 (default 20)

Model object

json

{
  "id": "mdl_01HZX9VABCDEF",
  "slug": "acme/grasp-policy-v2",
  "name": "Grasp Policy v2",
  "owner": { "id": "usr_01HZ...", "slug": "acme" },
  "task": "manipulation",
  "framework": "pytorch",
  "license": "MIT",
  "price": { "currency": "usd", "amount": 4900, "model": "one_time" },
  "latest_revision": "1.4.0",
  "revisions": ["1.0.0", "1.1.0", "1.2.0", "1.3.0", "1.4.0"],
  "downloads": 12_487,
  "stars": 421,
  "created_at": "2026-03-19T18:04:11Z",
  "updated_at": "2026-05-22T09:11:47Z"
}

Retrieve a model

http

GET /v1/models/{id_or_slug}
GET /v1/models/{id_or_slug}/revisions/{revision}

Download a model

http

GET /v1/models/{id}/download
Authorization: Bearer <key>

Response — 302
  Location: https://cdn.nodedata.dev/.../weights.safetensors?signature=...

Region-aware

Download URLs are signed and routed to the nearest CDN edge. Links expire after 30 minutes; refresh as needed by calling the endpoint again.

Datasets

The dataset endpoints mirror models. See:

GET /v1/datasets
POST /v1/datasets
POST /v1/datasets/{id}/shards — resumable upload
GET /v1/datasets/{id}/shards/{name}
POST /v1/datasets/{id}/publish

Listings

POST /v1/listings — create a storefront for an asset
PATCH /v1/listings/{id} — update pricing, title, gallery
POST /v1/listings/{id}/publish
POST /v1/listings/{id}/unpublish

Purchases

http

POST /v1/purchases
Authorization: Bearer <key>
Content-Type: application/json

{
  "listing_id": "lst_01HZ...",
  "payment_method": "pm_01HZ..."   // or "balance"
}

Response — 201 Created
{
  "id": "pur_01HZ...",
  "listing_id": "lst_01HZ...",
  "status": "succeeded",
  "amount": { "currency": "usd", "amount": 4900 },
  "platform_fee": { "currency": "usd", "amount": 1225 },
  "seller_payout": { "currency": "usd", "amount": 3675 },
  "license_grant_id": "grt_01HZ..."
}

Deployments

POST /v1/deployments — create a hosted endpoint from a model
GET /v1/deployments/{id}/logs
POST /v1/deployments/{id}/invoke — run inference

Webhooks

POST /v1/webhook_endpoints
GET /v1/webhook_endpoints/{id}/events — replay log

See Authentication for signature verification.

Develop

API reference

Every endpoint, every parameter, and every response shape — typed and versioned.

Base URL

text

https://api.nodedata.dev/v1

All endpoints accept JSON bodies and return JSON responses. Errors follow the same shape across the API.

Versioning

Error format

json

{
  "error": {
    "code": "validation_failed",
    "message": "license must be one of MIT, Apache-2.0, CC-BY-4.0",
    "field": "license",
    "request_id": "req_01HZ..."
  }
}

Always log request_id. Support requests resolve much faster when the request ID is included.

Pagination

List endpoints return a cursor in next_cursor when more results exist. Pass it back as ?cursor= on the next call.

json

{
  "items": [/* ... */],
  "next_cursor": "eyJsYXN0X2lkIjoibWRsXzAxSF...",
  "has_more": true
}

Rate limits

Surface	Limit
Read endpoints	1,000 req / minute / key
Write endpoints	120 req / minute / key
Upload endpoints	30 req / minute / key (uploads themselves are unmetered)
Inference (hosted)	Per listing — see the listing's gateway docs

Every response carries X-RateLimit-Remaining and X-RateLimit-Reset headers. When throttled the API responds with 429 and a Retry-After header in seconds.

Account

Get the authenticated account

http

GET /v1/me
Authorization: Bearer <key>

Response — 200
{
  "id": "usr_01HZ...",
  "slug": "acme",
  "name": "ACME Robotics",
  "email": "team@acme.dev",
  "created_at": "2026-03-19T18:04:11Z",
  "key": { "mode": "test", "scopes": ["models:read"] }
}

Smallest possible call to verify a key. Returns the account that owns the key plus the key's mode and scopes.

Models

List models

http

GET /v1/models
Authorization: Bearer <key>

Query parameters
  q          string   Full-text search query
  task       enum     manipulation | navigation | perception | language | vla
  framework  enum     pytorch | tensorflow | jax | onnx | gguf
  license    enum     MIT | Apache-2.0 | CC-BY-4.0 | Commercial
  cursor     string   Pagination cursor returned in the previous response
  limit      int      1–100 (default 20)

Model object

json

{
  "id": "mdl_01HZX9VABCDEF",
  "slug": "acme/grasp-policy-v2",
  "name": "Grasp Policy v2",
  "owner": { "id": "usr_01HZ...", "slug": "acme" },
  "task": "manipulation",
  "framework": "pytorch",
  "license": "MIT",
  "price": { "currency": "usd", "amount": 4900, "model": "one_time" },
  "latest_revision": "1.4.0",
  "revisions": ["1.0.0", "1.1.0", "1.2.0", "1.3.0", "1.4.0"],
  "downloads": 12_487,
  "stars": 421,
  "created_at": "2026-03-19T18:04:11Z",
  "updated_at": "2026-05-22T09:11:47Z"
}

Retrieve a model

http

GET /v1/models/{id_or_slug}
GET /v1/models/{id_or_slug}/revisions/{revision}

Download a model

http

GET /v1/models/{id}/download
Authorization: Bearer <key>

Response — 302
  Location: https://cdn.nodedata.dev/.../weights.safetensors?signature=...

Region-aware

Download URLs are signed and routed to the nearest CDN edge. Links expire after 30 minutes; refresh as needed by calling the endpoint again.

Datasets

The dataset endpoints mirror models. See:

GET /v1/datasets
POST /v1/datasets
POST /v1/datasets/{id}/shards — resumable upload
GET /v1/datasets/{id}/shards/{name}
POST /v1/datasets/{id}/publish

Listings

POST /v1/listings — create a storefront for an asset
PATCH /v1/listings/{id} — update pricing, title, gallery
POST /v1/listings/{id}/publish
POST /v1/listings/{id}/unpublish

Purchases

http

POST /v1/purchases
Authorization: Bearer <key>
Content-Type: application/json

{
  "listing_id": "lst_01HZ...",
  "payment_method": "pm_01HZ..."   // or "balance"
}

Response — 201 Created
{
  "id": "pur_01HZ...",
  "listing_id": "lst_01HZ...",
  "status": "succeeded",
  "amount": { "currency": "usd", "amount": 4900 },
  "platform_fee": { "currency": "usd", "amount": 1225 },
  "seller_payout": { "currency": "usd", "amount": 3675 },
  "license_grant_id": "grt_01HZ..."
}

Deployments

POST /v1/deployments — create a hosted endpoint from a model
GET /v1/deployments/{id}/logs
POST /v1/deployments/{id}/invoke — run inference

Webhooks

POST /v1/webhook_endpoints
GET /v1/webhook_endpoints/{id}/events — replay log

See Authentication for signature verification.