Skip to content

Commit e73c23b

Browse files
szabosteveszabosteve
authored
[8.x][DOCS] Adds VoyageAI inference integration docs. (elastic#125196)
1 parent 5b38a54 commit e73c23b

File tree

3 files changed

+180
-0
lines changed

3 files changed

+180
-0
lines changed

‎docs/reference/inference/inference-apis.asciidoc‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -151,4 +151,5 @@ include::service-hugging-face.asciidoc[]
151151
include::service-jinaai.asciidoc[]
152152
include::service-mistral.asciidoc[]
153153
include::service-openai.asciidoc[]
154+
include::service-voyageai.asciidoc[]
154155
include::service-watsonx-ai.asciidoc[]

‎docs/reference/inference/put-inference.asciidoc‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -78,6 +78,7 @@ Click the links to review the configuration details of the integrations:
7878
* <<infer-service-hugging-face,Hugging Face>> (`text_embedding`)
7979
* <<infer-service-mistral,Mistral>> (`text_embedding`)
8080
* <<infer-service-openai,OpenAI>> (`chat_completion`, `completion`, `text_embedding`)
81+
* <<infer-service-voyageai,VoyageAI>> (`text_embedding`, `rerank`)
8182
* <<infer-service-watsonx-ai>> (`text_embedding`)
8283
* <<infer-service-jinaai,JinaAI>> (`text_embedding`, `rerank`)
8384

‎docs/reference/inference/service-voyageai.asciidoc‎

Lines changed: 178 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,178 @@
1+
[[infer-service-voyageai]]
2+
=== VoyageAI {infer} integration
3+
4+
.New API reference
5+
[sidebar]
6+
--
7+
For the most up-to-date API details, refer to {api-es}/group/endpoint-inference[{infer-cap} APIs].
8+
--
9+
10+
Creates an {infer} endpoint to perform an {infer} task with the `voyageai` service.
11+
12+
13+
[discrete]
14+
[[infer-service-voyageai-api-request]]
15+
==== {api-request-title}
16+
17+
`PUT /_inference/<task_type>/<inference_id>`
18+
19+
[discrete]
20+
[[infer-service-voyageai-api-path-params]]
21+
==== {api-path-parms-title}
22+
23+
`<inference_id>`::
24+
(Required, string)
25+
include::inference-shared.asciidoc[tag=inference-id]
26+
27+
`<task_type>`::
28+
(Required, string)
29+
include::inference-shared.asciidoc[tag=task-type]
30+
+
31+
--
32+
Available task types:
33+
34+
* `text_embedding`,
35+
* `rerank`.
36+
--
37+
38+
[discrete]
39+
[[infer-service-voyageai-api-request-body]]
40+
==== {api-request-body-title}
41+
42+
`chunking_settings`::
43+
(Optional, object)
44+
include::inference-shared.asciidoc[tag=chunking-settings]
45+
46+
`max_chunk_size`:::
47+
(Optional, integer)
48+
include::inference-shared.asciidoc[tag=chunking-settings-max-chunking-size]
49+
50+
`overlap`:::
51+
(Optional, integer)
52+
include::inference-shared.asciidoc[tag=chunking-settings-overlap]
53+
54+
`sentence_overlap`:::
55+
(Optional, integer)
56+
include::inference-shared.asciidoc[tag=chunking-settings-sentence-overlap]
57+
58+
`strategy`:::
59+
(Optional, string)
60+
include::inference-shared.asciidoc[tag=chunking-settings-strategy]
61+
62+
`service`::
63+
(Required, string)
64+
The type of service supported for the specified task type. In this case,
65+
`voyageai`.
66+
67+
`service_settings`::
68+
(Required, object)
69+
include::inference-shared.asciidoc[tag=service-settings]
70+
+
71+
--
72+
These settings are specific to the `voyageai` service.
73+
--
74+
75+
`dimensions`:::
76+
(Optional, integer)
77+
The number of dimensions the resulting output embeddings should have.
78+
This setting maps to `output_dimension` in the https://docs.voyageai.com/docs/embeddings[VoyageAI documentation].
79+
Only for the `text_embedding` task type.
80+
81+
`embedding_type`:::
82+
(Optional, string)
83+
The data type for the embeddings to be returned.
84+
This setting maps to `output_dtype` in the https://docs.voyageai.com/docs/embeddings[VoyageAI documentation].
85+
Permitted values: `float`, `int8`, `bit`.
86+
`int8` is a synonym of `byte` in the VoyageAI documentation.
87+
`bit` is a synonym of `binary` in the VoyageAI documentation.
88+
Only for the `text_embedding` task type.
89+
90+
`model_id`:::
91+
(Required, string)
92+
The name of the model to use for the {infer} task.
93+
Refer to the VoyageAI documentation for the list of available https://docs.voyageai.com/docs/embeddings[text embedding] and https://docs.voyageai.com/docs/reranker[rerank] models.
94+
95+
`rate_limit`:::
96+
(Optional, object)
97+
This setting helps to minimize the number of rate limit errors returned from VoyageAI.
98+
The `voyageai` service sets a default number of requests allowed per minute depending on the task type.
99+
For both `text_embedding` and `rerank`, it is set to `2000`.
100+
To modify this, set the `requests_per_minute` setting of this object in your service settings:
101+
+
102+
--
103+
include::inference-shared.asciidoc[tag=request-per-minute-example]
104+
105+
More information about the rate limits for OpenAI can be found in your https://platform.openai.com/account/limits[Account limits].
106+
--
107+
108+
`task_settings`::
109+
(Optional, object)
110+
include::inference-shared.asciidoc[tag=task-settings]
111+
+
112+
.`task_settings` for the `text_embedding` task type
113+
[%collapsible%closed]
114+
=====
115+
`input_type`:::
116+
(Optional, string)
117+
Type of the input text.
118+
Permitted values: `ingest` (maps to `document` in the VoyageAI documentation), `search` (maps to `query` in the VoyageAI documentation).
119+
120+
`truncation`:::
121+
(Optional, boolean)
122+
Whether to truncate the input texts to fit within the context length.
123+
Defaults to `false`.
124+
=====
125+
+
126+
.`task_settings` for the `rerank` task type
127+
[%collapsible%closed]
128+
=====
129+
`return_documents`:::
130+
(Optional, boolean)
131+
Whether to return the source documents in the response.
132+
Defaults to `false`.
133+
134+
`top_k`:::
135+
(Optional, integer)
136+
The number of most relevant documents to return.
137+
If not specified, the reranking results of all documents will be returned.
138+
139+
`truncation`:::
140+
(Optional, boolean)
141+
Whether to truncate the input texts to fit within the context length.
142+
Defaults to `false`.
143+
=====
144+
145+
146+
[discrete]
147+
[[inference-example-voyageai]]
148+
==== VoyageAI service example
149+
150+
The following example shows how to create an {infer} endpoint called `voyageai-embeddings` to perform a `text_embedding` task type.
151+
The embeddings created by requests to this endpoint will have 512 dimensions.
152+
153+
[source,console]
154+
------------------------------------------------------------
155+
PUT _inference/text_embedding/voyageai-embeddings
156+
{
157+
"service": "voyageai",
158+
"service_settings": {
159+
"model_id": "voyage-3-large",
160+
"dimensions": 512
161+
}
162+
}
163+
------------------------------------------------------------
164+
// TEST[skip:TBD]
165+
166+
The next example shows how to create an {infer} endpoint called `voyageai-rerank` to perform a `rerank` task type.
167+
168+
[source,console]
169+
------------------------------------------------------------
170+
PUT _inference/rerank/voyageai-rerank
171+
{
172+
"service": "voyageai",
173+
"service_settings": {
174+
"model_id": "rerank-2"
175+
}
176+
}
177+
------------------------------------------------------------
178+
// TEST[skip:TBD]

0 commit comments

Comments
 (0)