async_openai/types/embeddings/
embedding.rs

1use derive_builder::Builder;
2use serde::{Deserialize, Serialize};
3
4use crate::error::OpenAIError;
5
6#[derive(Debug, Serialize, Clone, PartialEq, Deserialize)]
7#[serde(untagged)]
8pub enum EmbeddingInput {
9    String(String),
10    StringArray(Vec<String>),
11    // Minimum value is 0, maximum value is 100257 (inclusive).
12    IntegerArray(Vec<u32>),
13    ArrayOfIntegerArray(Vec<Vec<u32>>),
14}
15
16#[derive(Debug, Serialize, Default, Clone, PartialEq, Deserialize)]
17#[serde(rename_all = "lowercase")]
18pub enum EncodingFormat {
19    #[default]
20    Float,
21    Base64,
22}
23
24#[derive(Debug, Serialize, Default, Clone, Builder, PartialEq, Deserialize)]
25#[builder(name = "CreateEmbeddingRequestArgs")]
26#[builder(pattern = "mutable")]
27#[builder(setter(into, strip_option), default)]
28#[builder(derive(Debug))]
29#[builder(build_fn(error = "OpenAIError"))]
30pub struct CreateEmbeddingRequest {
31    /// ID of the model to use. You can use the [List models](https://platform.openai.com/docs/api-reference/models/list)
32    /// API to see all of your available models, or see our [Model overview](https://platform.openai.com/docs/models)
33    /// for descriptions of them.
34    pub model: String,
35
36    /// Input text to embed, encoded as a string or array of tokens. To embed multiple inputs in a single
37    /// request, pass an array of strings or array of token arrays. The input must not exceed the max
38    /// input tokens for the model (8192 tokens for all embedding models), cannot be an empty string, and
39    /// any array must be 2048 dimensions or less. [Example Python
40    /// code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken) for counting tokens.
41    /// In addition to the per-input token limit, all embedding  models enforce a maximum of 300,000
42    /// tokens summed across all inputs in a  single request.
43    pub input: EmbeddingInput,
44
45    /// The format to return the embeddings in. Can be either `float` or [`base64`](https://pypi.org/project/pybase64/).
46    #[serde(skip_serializing_if = "Option::is_none")]
47    pub encoding_format: Option<EncodingFormat>,
48
49    /// A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse.
50    /// [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
51    #[serde(skip_serializing_if = "Option::is_none")]
52    pub user: Option<String>,
53
54    /// The number of dimensions the resulting output embeddings should have. Only supported in `text-embedding-3` and later models.
55    #[serde(skip_serializing_if = "Option::is_none")]
56    pub dimensions: Option<u32>,
57}
58
59/// Represents an embedding vector returned by embedding endpoint.
60#[derive(Debug, Deserialize, Serialize, Clone, PartialEq, Default)]
61pub struct Embedding {
62    /// The index of the embedding in the list of embeddings.
63    pub index: u32,
64    /// The object type, which is always "embedding".
65    pub object: String,
66    /// The embedding vector, returned as a list of floats when `encoding_format` is `float` (the default),
67    /// or as a base64-encoded string when `encoding_format` is `base64`. The length of the vector depends
68    /// on the model as listed in the [embedding
69    /// guide](https://developers.openai.com/api/docs/guides/embeddings).
70    pub embedding: Vec<f32>,
71}
72
73#[derive(Debug, Deserialize, Serialize, Clone, PartialEq, Default)]
74pub struct Base64EmbeddingVector(pub String);
75
76/// Represents an base64-encoded embedding vector returned by embedding endpoint.
77#[derive(Debug, Deserialize, Serialize, Clone, PartialEq, Default)]
78pub struct Base64Embedding {
79    /// The index of the embedding in the list of embeddings.
80    pub index: u32,
81    /// The object type, which is always "embedding".
82    pub object: String,
83    /// The embedding vector, encoded in base64.
84    pub embedding: Base64EmbeddingVector,
85}
86
87#[derive(Debug, Deserialize, Serialize, Clone, PartialEq, Default)]
88pub struct EmbeddingUsage {
89    /// The number of tokens used by the prompt.
90    pub prompt_tokens: u32,
91    /// The total number of tokens used by the request.
92    pub total_tokens: u32,
93}
94
95#[derive(Debug, Deserialize, Clone, PartialEq, Serialize, Default)]
96pub struct CreateEmbeddingResponse {
97    pub object: String,
98    /// The name of the model used to generate the embedding.
99    pub model: String,
100    /// The list of embeddings generated by the model.
101    pub data: Vec<Embedding>,
102    /// The usage information for the request.
103    pub usage: EmbeddingUsage,
104}
105
106#[derive(Debug, Deserialize, Clone, PartialEq, Serialize, Default)]
107pub struct CreateBase64EmbeddingResponse {
108    pub object: String,
109    /// The name of the model used to generate the embedding.
110    pub model: String,
111    /// The list of embeddings generated by the model.
112    pub data: Vec<Base64Embedding>,
113    /// The usage information for the request.
114    pub usage: EmbeddingUsage,
115}
async_openai/types/embeddings/embedding.rs

async_openai/types/embeddings/
embedding.rs
