You're looking at a specific version of this model. Jump to the model overview.
Input schema
The fields you can use to run this model with an API. If you don’t give a value for a field its default value will be used.
| Field | Type | Default value | Description |
|---|---|---|---|
| texts |
string
|
A JSON-formatted list of strings to encode.
|
|
| options |
string
|
A JSON-formatted string containing additional options for encoding.{ 'for_retrieval': bool //set to true to query embeddings suitable for document retrieval, false for storage embeddings. (default: false), 'normalize': bool //set to true to normalize returned embedding vectors to a magnitude of 1. (default: true, as this model presumes cosine similarity comparisons downstream), 'batchtoken_max': int //optional, maximumum number of tokens to try to stuff into a batch (to avoid out of memory errors). If the total number of tokens across the list of requested embeddings exceed this value, list will be split internally and run across multiple forward passes. The result will still be a single list. for a 40GB GPU, a good numnber is 16384 'precision' : str //optional, either full or half. defaults to a paranoid value of full}
|
Output schema
The shape of the response you’ll get when you run this model with an API.
Schema
{'properties': {'embeddings': {'items': {'items': {'type': 'number'},
'type': 'array'},
'title': 'Embeddings',
'type': 'array'},
'metrics': {'title': 'Metrics', 'type': 'string'}},
'required': ['embeddings', 'metrics'],
'title': 'Output',
'type': 'object'}