Inference API

Embeddings

POST /v1/embeddings

Create an embedding vector representation corresponding to the input text. This is the endpoint for making requests to embedding models.

Request Body

dimensions (integer | null) — The number of dimensions the resulting output embeddings should have.
encoding_format (string | null) — The format to return the embeddings in. Can be either `float` or `base64`.
input (object | object | object | object)
- String (string, required) — A strings to be embedded. For best performance, prepend "query: " in front of query content and prepend "passage: " in front of passage/text
- StringArray (array<string>, required) — An array of strings to be embedded
- Ints (array<integer>, required) — A token in integer to be embedded
- IntsArray (array<array<integer>>, required) — An array of tokens in integers to be embedded
model (string) — ID of the model to use.
preview (boolean | null) — Flag to use the new format of the API.
user (string | null) — A unique identifier representing your end-user, which can help xAI to monitor and detect abuse.

Response Body

data (array<object>, required) — A list of embedding objects.
- embedding (string | array<number>, required)
- index (integer, required) — Index of the embedding object in the data list.
- object (string, required) — The object type, which is always `"embedding"`.
model (string, required) — Model ID used to create embedding.
object (string, required) — The object type of `data` field, which is always `"list"`.
usage (object)
- prompt_tokens (integer, required) — Prompt token used.
- total_tokens (integer, required) — Total token used.

**Request example:**

"{\n            \"input\": [\"This is an example content to embed...\"],\n            \"model\": \"v1\",\n            \"encoding_format\": \"float\"\n        }"

**Response example:**

{
  "object": "list",
  "model": "v1",
  "data": [
    {
      "index": 0,
      "embedding": [
        0.01567895,
        0.063257694,
        0.045925662
      ],
      "object": "embedding"
    }
  ],
  "usage": {
    "prompt_tokens": 1,
    "total_tokens": 1
  }
}

GET /v1/embedding-models

List all embedding models available to the authenticating API key with full information. Additional information compared to /v1/models includes modalities, fingerprint and alias(es).

Response Body

models (array<object>, required) — Array of available embedding models.
- aliases (array<string>, required) — Alias ID(s) of the model that user can use in a request's model field.
- created (integer, required) — Model creation time in Unix timestamp.
- fingerprint (string, required) — Fingerprint of the xAI system configuration hosting the model.
- id (string, required) — Model ID. Obtainable from <https://console.x.ai/team/default/models> or <https://docs.x.ai/docs/models>.
- input_modalities (array<string>, required) — The input modalities supported by the model.
- object (string, required) — Object type, should be model.
- output_modalities (array<string>, required) — The output modalities supported by the model.
- owned_by (string, required) — Owner of the model.
- prompt_image_token_price (integer, required) — Price of the prompt image token in USD cents per million token.
- prompt_text_token_price (integer, required) — Price of the prompt text token in USD cents per million token.
- version (string, required) — Version of the model.

**Response example:**

{
  "models": [
    {
      "id": "v1",
      "fingerprint": "fp_df37966059",
      "created": 1725148800,
      "object": "model",
      "owned_by": "xai",
      "version": "0.1.0",
      "input_modalities": [
        "text"
      ],
      "prompt_text_token_price": 100,
      "prompt_image_token_price": 0,
      "aliases": []
    }
  ]
}

GET /v1/embedding-models/{model_id}

Get full information about an embedding model with its model_id.

Path Parameters

model_id (string, required) — ID of the model to get.

Response Body

aliases (array<string>, required) — Alias ID(s) of the model that user can use in a request's model field.
created (integer, required) — Model creation time in Unix timestamp.
fingerprint (string, required) — Fingerprint of the xAI system configuration hosting the model.
id (string, required) — Model ID. Obtainable from <https://console.x.ai/team/default/models> or <https://docs.x.ai/docs/models>.
input_modalities (array<string>, required) — The input modalities supported by the model.
object (string, required) — Object type, should be model.
output_modalities (array<string>, required) — The output modalities supported by the model.
owned_by (string, required) — Owner of the model.
prompt_image_token_price (integer, required) — Price of the prompt image token in USD cents per million token.
prompt_text_token_price (integer, required) — Price of the prompt text token in USD cents per million token.
version (string, required) — Version of the model.

**Response example:**

{
  "id": "v1",
  "created": 1725148800,
  "object": "model",
  "owned_by": "xai",
  "version": "0.1.0",
  "input_modalities": [
    "text"
  ],
  "prompt_text_token_price": 10,
  "prompt_image_token_price": 0,
  "aliases": []
}

rest-api-reference/inference/embeddings.md +188 −0 created

1#### Inference API

3# Embeddings

5***

7## POST /v1/embeddings

9Create an embedding vector representation corresponding to the input text. This is the endpoint for making requests to embedding models.

11### Request Body

13* `dimensions` (integer | null) — The number of dimensions the resulting output embeddings should have.

15* `encoding_format` (string | null) — The format to return the embeddings in. Can be either \`float\` or \`base64\`.

17* `input` (object | object | object | object)

19 * `String` (string, required) — A strings to be embedded. For best performance, prepend "query: " in front of query content and prepend "passage: " in front of passage/text

21 * `StringArray` (array\<string>, required) — An array of strings to be embedded

23 * `Ints` (array\<integer>, required) — A token in integer to be embedded

25 * `IntsArray` (array\<array\<integer>>, required) — An array of tokens in integers to be embedded

27* `model` (string) — ID of the model to use.

29* `preview` (boolean | null) — Flag to use the new format of the API.

31* `user` (string | null) — A unique identifier representing your end-user, which can help xAI to monitor and detect abuse.

33### Response Body

35* `data` (array\<object>, required) — A list of embedding objects.

37 * `embedding` (string | array\<number>, required)

39 * `index` (integer, required) — Index of the embedding object in the data list.

41 * `object` (string, required) — The object type, which is always \`"embedding"\`.

43* `model` (string, required) — Model ID used to create embedding.

45* `object` (string, required) — The object type of \`data\` field, which is always \`"list"\`.

47* `usage` (object)

49 * `prompt_tokens` (integer, required) — Prompt token used.

51 * `total_tokens` (integer, required) — Total token used.

53\*\*Request example:\*\*

55```json

56"{\n \"input\": [\"This is an example content to embed...\"],\n \"model\": \"v1\",\n \"encoding_format\": \"float\"\n }"

57```

59\*\*Response example:\*\*

61```json

62{

63 "object": "list",

64 "model": "v1",

65 "data": [

66 {

67 "index": 0,

68 "embedding": [

69 0.01567895,

70 0.063257694,

71 0.045925662

72 ],

73 "object": "embedding"

74 }

75 ],

76 "usage": {

77 "prompt_tokens": 1,

78 "total_tokens": 1

79 }

80}

81```

83***

85## GET /v1/embedding-models

87List all embedding models available to the authenticating API key with full information. Additional information compared to /v1/models includes modalities, fingerprint and alias(es).

89### Response Body

91* `models` (array\<object>, required) — Array of available embedding models.

93 * `aliases` (array\<string>, required) — Alias ID(s) of the model that user can use in a request's model field.

95 * `created` (integer, required) — Model creation time in Unix timestamp.

97 * `fingerprint` (string, required) — Fingerprint of the xAI system configuration hosting the model.

99 * `id` (string, required) — Model ID. Obtainable from \<https://console.x.ai/team/default/models> or \<https://docs.x.ai/docs/models>.

100

101 * `input_modalities` (array\<string>, required) — The input modalities supported by the model.

102

103 * `object` (string, required) — Object type, should be model.

104

105 * `output_modalities` (array\<string>, required) — The output modalities supported by the model.

106

107 * `owned_by` (string, required) — Owner of the model.

108

109 * `prompt_image_token_price` (integer, required) — Price of the prompt image token in USD cents per million token.

110

111 * `prompt_text_token_price` (integer, required) — Price of the prompt text token in USD cents per million token.

112

113 * `version` (string, required) — Version of the model.

114

115\*\*Response example:\*\*

116

117```json

118{

119 "models": [

120 {

121 "id": "v1",

122 "fingerprint": "fp_df37966059",

123 "created": 1725148800,

124 "object": "model",

125 "owned_by": "xai",

126 "version": "0.1.0",

127 "input_modalities": [

128 "text"

129 ],

130 "prompt_text_token_price": 100,

131 "prompt_image_token_price": 0,

132 "aliases": []

133 }

134 ]

135}

136```

137

138***

139

140## GET /v1/embedding-models/\{model\_id}

141

142Get full information about an embedding model with its model\_id.

143

144### Path Parameters

145

146* `model_id` (string, required) — ID of the model to get.

147

148### Response Body

149

150* `aliases` (array\<string>, required) — Alias ID(s) of the model that user can use in a request's model field.

151

152* `created` (integer, required) — Model creation time in Unix timestamp.

153

154* `fingerprint` (string, required) — Fingerprint of the xAI system configuration hosting the model.

155

156* `id` (string, required) — Model ID. Obtainable from \<https://console.x.ai/team/default/models> or \<https://docs.x.ai/docs/models>.

157

158* `input_modalities` (array\<string>, required) — The input modalities supported by the model.

159

160* `object` (string, required) — Object type, should be model.

161

162* `output_modalities` (array\<string>, required) — The output modalities supported by the model.

163

164* `owned_by` (string, required) — Owner of the model.

165

166* `prompt_image_token_price` (integer, required) — Price of the prompt image token in USD cents per million token.

167

168* `prompt_text_token_price` (integer, required) — Price of the prompt text token in USD cents per million token.

169

170* `version` (string, required) — Version of the model.

171

172\*\*Response example:\*\*

173

174```json

175{

176 "id": "v1",

177 "created": 1725148800,

178 "object": "model",

179 "owned_by": "xai",

180 "version": "0.1.0",

181 "input_modalities": [

182 "text"

183 ],

184 "prompt_text_token_price": 10,

185 "prompt_image_token_price": 0,

186 "aliases": []

187}

188```

rest-api-reference/inference/embeddings.md 2026-09-03 16:59 UTC to 2026-09-07 22:57 UTC

Inference API

Embeddings

POST /v1/embeddings

Request Body

Response Body

GET /v1/embedding-models

Response Body

GET /v1/embedding-models/{model_id}

Path Parameters

Response Body