GET

/v1/schemas/nvidia/embeddings/nvidia%2Fnemotron-3.5-embed-1b

kind inputactivity embeddingshash 1fe8bb86…extracted Oct 21, 2026version currentsource upstream spec ↗

Request body

10 properties · 2 required · 0 $defs
propertytypedescriptionconstraints
input*string | string[]Input to embed. Each item is text, one image, or text followed by one image. Send an image as a base64 data URL, for example `data:image/png;base64,<data>`, or as an `<img src="data:image/png;base64,<data>">` tag. PNG and JPEG images are supported, with one im…length ≥ 1
model*stringID of the embedding model.default: "nvidia/nemotron-3.5-embed-1b"
input_typeenumnvidia/nemotron-3.5-embed-1b operates in `passage` or `query` mode, and thus requires the `input_type` parameter. `passage` is used when generating embeddings during indexing. `query` is used when generating embeddings during querying. Inputs that contain an i…passage | query
modalityenumThe modality of each input item: `text`, `image`, or `text_image`. Give one value for all items, or one value per item. If omitted, the modality is inferred from the input.text | image | text_image
encoding_formatenumThe format to return the embeddings in.float | base64 · default: "float"
truncateenumSpecifies how inputs longer than the maximum token length (10240) are handled. `END` discards the end of the text until the input fits. `NONE` returns an error when the input exceeds the maximum. `START` is not supported. The image and the passage prefix must …NONE | END · default: "END"
dimensionsintegerNumber of dimensions of the returned embedding, from 128 through the native dimension of 2048. Shortened vectors are L2-normalized again. Omit to receive the native 2048-dimensional embedding.128 ≤ n ≤ 2048
embedding_typeenumOutput representation of each embedding value. Quantized representations pack several dimensions into each output element.float | int8 | uint8 | binary | ubinary · default: "float"
longest_edgeintegerMaximum image size before patch-grid alignment. Larger images are downsized while preserving the aspect ratio; smaller images are not enlarged. Omit to use the model default of 1120. Use the same value when indexing and querying.256 ≤ n ≤ 1540
userstringOptional identifier for the end user. If provided it must be non-blank and at most 512 characters. It does not select the model or authorize the request.length ≤ 512

Validate a payload

before you spend tokens
shell
$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"nvidia","endpointId":"nvidia/nemotron-3.5-embed-1b","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}