Skip to content

Commit 93ed30c

Browse files
jborsecnikjborsecnik
authored
Merge pull request #189789 from aahill/limits-article
Language service limits
2 parents 43b357d + 4495797 commit 93ed30c

File tree

15 files changed

+153
-243
lines changed

15 files changed

+153
-243
lines changed

‎articles/cognitive-services/language-service/concepts/data-limits.md

Lines changed: 100 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,100 @@
1+
---
2+
title: Data limits for Language service features
3+
titleSuffix: Azure Cognitive Services
4+
description: Data and service limitations for Azure Cognitive Service for Language features.
5+
services: cognitive-services
6+
author: aahill
7+
manager: nitinme
8+
ms.service: cognitive-services
9+
ms.subservice: language-service
10+
ms.topic: conceptual
11+
ms.date: 02/25/2022
12+
ms.author: aahi
13+
---
14+
15+
# Service limits for Azure Cognitive Service for Language
16+
17+
> [!NOTE]
18+
> This article only describes the limits for pre-configured features in Azure Cognitive Service for Language:
19+
> To see the service limits for customizable features, see the following articles:
20+
> * [Custom classification](../custom-classification/service-limits.md)
21+
> * [Custom NER](../custom-named-entity-recognition/service-limits.md)
22+
> * [Conversational language understanding](../conversational-language-understanding/service-limits.md)
23+
> * [Question answering](../question-answering/concepts/limits.md)
24+
25+
Use this article to find the limits for the size, and rates that you can send data to the following features of the language service.
26+
* [Named Entity Recognition (NER)](../named-entity-recognition/overview.md)
27+
* [Personally Identifiable Information (PII) detection](../personally-identifiable-information/overview.md)
28+
* [Key phrase extraction](../key-phrase-extraction/overview.md)
29+
* [Entity linking](../entity-linking/overview.md)
30+
* [Text Analytics for health](../text-analytics-for-health/overview.md)
31+
* [Sentiment analysis and opinion mining](../sentiment-opinion-mining/overview.md)
32+
* [Language detection](../language-detection/overview.md)
33+
34+
When using features of the Language service, keep the following in mind:
35+
36+
* Pricing is not affected by data or rate limits. Pricing is based on the number of text records you send to the API, and is subject to your Language resource's [pricing details](https://aka.ms/unifiedLanguagePricing).
37+
* A text record is measured as 1000 characters.
38+
* Data and rate limits are based on the number of documents you send to the API. If you need to analyze larger documents than the limit allows, you can break the text into smaller chunks of text before sending them to the API.
39+
* A document is a single string of text characters.
40+
41+
## Maximum characters per document
42+
43+
The following limit specifies the maximum number of characters that can be in a single document.
44+
45+
| Feature | Value |
46+
|------------------------|---------------|
47+
| Text Analytics for health | 30,720 characters as measured by [StringInfo.LengthInTextElements](/dotnet/api/system.globalization.stringinfo.lengthintextelements). |
48+
| All other pre-configured features (synchronous) | 5,120 as measured by [StringInfo.LengthInTextElements](/dotnet/api/system.globalization.stringinfo.lengthintextelements). |
49+
| All other pre-configured features ([asynchronous](use-asynchronously.md)) | 125,000 characters across all submitted documents, as measured by [StringInfo.LengthInTextElements](/dotnet/api/system.globalization.stringinfo.lengthintextelements) (maximum of 25 documents). |
50+
51+
If a document exceeds the character limit, the API will behave differently depending on how you're sending requests.
52+
53+
If you're sending requests synchronously:
54+
* The API won't process a document that exceeds the maximum size, and will return an invalid document error for it. If an API request has multiple documents, the API will continue processing them if they are within the character limit.
55+
56+
If you're sending requests [asynchronously](use-asynchronously.md):
57+
* The API will reject the entire request and return a `400 bad request` error if any document within it exceeds the maximum size.
58+
59+
## Maximum request size
60+
61+
The following limit specifies the maximum size of documents contained in the entire request.
62+
63+
| Feature | Value |
64+
|------------------------|---------------|
65+
| All pre-configured features | 1MB |
66+
67+
## Maximum documents per request
68+
69+
Exceeding the following document limits will generate an HTTP 400 error code.
70+
71+
> [!NOTE]
72+
> When sending asynchronous API requests, you can send a maximum of 25 documents per request.
73+
74+
| Feature | Max Documents Per Request |
75+
|----------|-----------|
76+
| Language Detection | 1000 |
77+
| Sentiment Analysis | 10 |
78+
| Opinion Mining | 10 |
79+
| Key Phrase Extraction | 10 |
80+
| Named Entity Recognition (NER) | 5 |
81+
| Personally Identifying Information (PII) detection | 5 |
82+
| Text summarization | 25 |
83+
| Entity Linking | 5 |
84+
| Text Analytics for health | 10 for the web-based API, 1000 for the container. |
85+
86+
## Rate limits
87+
88+
Your rate limit will vary with your [pricing tier](https://azure.microsoft.com/pricing/details/cognitive-services/text-analytics/). These limits are the same for both versions of the API. These rate limits don't apply to the Text Analytics for health container, which does not have a set rate limit.
89+
90+
| Tier | Requests per second | Requests per minute |
91+
|---------------|---------------------|---------------------|
92+
| S / Multi-service | 1000 | 1000 |
93+
| S0 / F0 | 100 | 300 |
94+
95+
Requests rates are measured for each feature separately. You can send the maximum number of requests for your pricing tier to each feature, at the same time. For example, if you're in the `S` tier and send 1000 requests at once, you wouldn't be able to send another request for 59 seconds.
96+
97+
## See also
98+
99+
* [What is Azure Cognitive Service for Language](../overview.md)
100+
* [Pricing details](https://aka.ms/unifiedLanguagePricing)

‎articles/cognitive-services/language-service/entity-linking/how-to/call-api.md

Lines changed: 3 additions & 28 deletions
Original file line numberDiff line numberDiff line change
@@ -39,42 +39,17 @@ To send an API request, You will need a Language resource endpoint and key.
3939
> [!NOTE]
4040
> You can find the key and endpoint for your Language resource on the Azure portal. They will be located on the resource's **Key and endpoint** page, under **resource management**.
4141
42-
Analysis is performed upon receipt of the request. For information on the size and number of requests you can send per minute and second, see the data limits below.
43-
44-
Using entity linking synchronously is stateless. No data is stored in your account, and results are returned immediately in the response.
42+
Analysis is performed upon receipt of the request. Using entity linking synchronously is stateless. No data is stored in your account, and results are returned immediately in the response.
4543

4644
[!INCLUDE [asynchronous-result-availability](../../includes/async-result-availability.md)]
4745

4846
### Getting entity linking results
4947

5048
You can stream the results to an application, or save the output to a file on the local system.
5149

52-
## Data limits
53-
54-
> [!NOTE]
55-
> * If you need to analyze larger documents than the limit allows, you can break the text into smaller chunks of text before sending them to the API.
56-
> * A document is a single string of text characters.
57-
58-
| Limit | Value |
59-
|------------------------|---------------|
60-
| Maximum size of a single document | 5,120 characters as measured by [StringInfo.LengthInTextElements](/dotnet/api/system.globalization.stringinfo.lengthintextelements). |
61-
| Maximum number of characters per request (asynchronous) | 125K characters across all submitted documents, as measured by [StringInfo.LengthInTextElements](/dotnet/api/system.globalization.stringinfo.lengthintextelements). |
62-
| Maximum size of entire request | 1 MB |
63-
| Max Documents Per Request | 5 |
64-
65-
If a document exceeds the character limit, the API will behave differently depending on whether you're using it synchronously or asynchronously:
66-
67-
* Asynchronous: The API will reject the entire request and return a `400 bad request` error if any document within it exceeds the maximum size.
68-
* Synchronous: The API won't process a document that exceeds the maximum size, and will return an invalid document error for it. If an API request has multiple documents, the API will continue processing them if they are within the character limit.
69-
70-
### Rate limits
71-
72-
Your rate limit will vary with your [pricing tier](https://aka.ms/unifiedLanguagePricing).
50+
## Service and data limits
7351

74-
| Tier | Requests per second | Requests per minute |
75-
|---------------|---------------------|---------------------|
76-
| S / Multi-service | 1000 | 1000 |
77-
| S0 / F0 | 100 | 300 |
52+
[!INCLUDE [service limits article](../../includes/service-limits-link.md)]
7853

7954
## See also
8055

‎articles/cognitive-services/language-service/includes/service-limits-link.md

Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
---
2+
title: include file
3+
description: include file
4+
services: cognitive-services
5+
author: aahill
6+
ms.service: cognitive-services
7+
ms.topic: include
8+
ms.date: 12/02/2021
9+
ms.author: aahi
10+
ms.custom: include
11+
---
12+
13+
For information on the size and number of requests you can send per minute and second, see the [service limits](../concepts/data-limits.md) article.

‎articles/cognitive-services/language-service/key-phrase-extraction/how-to/call-api.md

Lines changed: 4 additions & 29 deletions
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@ manager: nitinme
88
ms.service: cognitive-services
99
ms.subservice: language-service
1010
ms.topic: how-to
11-
ms.date: 12/10/2021
11+
ms.date: 03/01/2022
1212
ms.author: aahi
1313
ms.custom: language-service-key-phrase, ignite-fall-2021
1414
---
@@ -42,9 +42,7 @@ To send an API request, You will need your Language resource endpoint and key.
4242
> [!NOTE]
4343
> You can find the key and endpoint for your Language resource on the Azure portal. They will be located on the resource's **Key and endpoint** page, under **resource management**.
4444
45-
Analysis is performed upon receipt of the request. For information on the size and number of requests you can send per minute and second, see the data limits below.
46-
47-
Using the key phrase extraction feature synchronously is stateless. No data is stored in your account, and results are returned immediately in the response.
45+
Analysis is performed upon receipt of the request. Using the key phrase extraction feature synchronously is stateless. No data is stored in your account, and results are returned immediately in the response.
4846

4947
[!INCLUDE [asynchronous-result-availability](../../includes/async-result-availability.md)]
5048

@@ -53,32 +51,9 @@ Using the key phrase extraction feature synchronously is stateless. No data is s
5351

5452
When you receive results from the API, the order of the returned key phrases is determined internally, by the model. You can stream the results to an application, or save the output to a file on the local system.
5553

56-
## Data limits
57-
58-
> [!NOTE]
59-
> * If you need to analyze larger documents than the limit allows, you can break the text into smaller chunks of text before sending them to the API.
60-
> * A document is a single string of text characters.
61-
62-
| Limit | Value |
63-
|------------------------|---------------|
64-
| Maximum size of a single document (synchronous) | 5,120 characters as measured by [StringInfo.LengthInTextElements](/dotnet/api/system.globalization.stringinfo.lengthintextelements). |
65-
| Maximum number of characters per request (asynchronous) | 125K characters across all submitted documents, as measured by [StringInfo.LengthInTextElements](/dotnet/api/system.globalization.stringinfo.lengthintextelements). |
66-
| Maximum size of entire request | 1 MB. |
67-
| Max Documents Per Request | 10 |
68-
69-
If a document exceeds the character limit, the API will behave differently depending on the endpoint you're using:
70-
71-
* Asynchronous: The API will reject the entire request and return a `400 bad request` error if any document within it exceeds the maximum size.
72-
* Synchronous: The API won't process a document that exceeds the maximum size, and will return an invalid document error for it. If an API request has multiple documents, the API will continue processing them if they are within the character limit.
73-
74-
### Rate limits
75-
76-
Your rate limit will vary with your [pricing tier](https://aka.ms/unifiedLanguagePricing).
54+
## Service and data limits
7755

78-
| Tier | Requests per second | Requests per minute |
79-
|---------------|---------------------|---------------------|
80-
| S / Multi-service | 1000 | 1000 |
81-
| S0 / F0 | 100 | 300 |
56+
[!INCLUDE [service limits article](../../includes/service-limits-link.md)]
8257

8358
## Next steps
8459

‎articles/cognitive-services/language-service/key-phrase-extraction/how-to/use-containers.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -21,7 +21,7 @@ Containers enable you to host the Key Phrase Extraction API on your own infrastr
2121

2222

2323
> [!NOTE]
24-
> * The free account is limited to 5,000 text records per month and only the **Free** and **Standard** [pricing tiers](https://azure.microsoft.com/pricing/details/cognitive-services/text-analytics) are valid for containers. For more information on transaction request rates, see [Data Limits](call-api.md#data-limits).
24+
> * The free account is limited to 5,000 text records per month and only the **Free** and **Standard** [pricing tiers](https://azure.microsoft.com/pricing/details/cognitive-services/text-analytics) are valid for containers. For more information on transaction request rates, see [Data and service limits](../../concepts/data-limits.md).
2525
2626
Containers enable you to run the Key Phrase Extraction APIs in your own environment and are great for your specific security and data governance requirements. The Key Phrase Extraction containers provide advanced natural language processing over raw text, and include three main functions: sentiment analysis, Key Phrase Extraction, and language detection.
2727

‎articles/cognitive-services/language-service/language-detection/how-to/call-api.md

Lines changed: 4 additions & 29 deletions
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@ manager: nitinme
88
ms.service: cognitive-services
99
ms.subservice: language-service
1010
ms.topic: how-to
11-
ms.date: 12/03/2021
11+
ms.date: 03/01/2022
1212
ms.author: aahi
1313
ms.custom: language-service-language-detection, ignite-fall-2021
1414
---
@@ -41,9 +41,7 @@ If you have content expressed in a less frequently used language, you can try th
4141
> [!TIP]
4242
> You can use a [Docker container](use-containers.md)for language detection, so you can use the API on-premises.
4343
44-
Analysis is performed upon receipt of the request. For information on the size and number of requests you can send per minute and second, see the data limits below.
45-
46-
Using the language detection feature synchronously is stateless. No data is stored in your account, and results are returned immediately in the response.
44+
Analysis is performed upon receipt of the request. Using the language detection feature synchronously is stateless. No data is stored in your account, and results are returned immediately in the response.
4745

4846
[!INCLUDE [asynchronous-result-availability](../../includes/async-result-availability.md)]
4947

@@ -174,32 +172,9 @@ The resulting output consists of the predominant language, with a score of less
174172
}
175173
```
176174

177-
## Data limits
178-
179-
> [!NOTE]
180-
> * If you need to analyze larger documents than the limit allows, you can break the text into smaller chunks of text before sending them to the API.
181-
> * A document is a single string of text characters.
182-
183-
| Limit | Value |
184-
|------------------------|---------------|
185-
| Maximum size of a single document (synchronous) | 5,120 characters as measured by [StringInfo.LengthInTextElements](/dotnet/api/system.globalization.stringinfo.lengthintextelements). |
186-
| Maximum number of characters per request (asynchronous) | 125K characters across all submitted documents, as measured by [StringInfo.LengthInTextElements](/dotnet/api/system.globalization.stringinfo.lengthintextelements). |
187-
| Maximum size of entire request | 1 MB. |
188-
| Max Documents Per Request | 1000 |
189-
190-
If a document exceeds the character limit, the API will behave differently depending on the endpoint you're using:
191-
192-
* Asynchronous: The API will reject the entire request and return a `400 bad request` error if any document within it exceeds the maximum size.
193-
* Synchronous: The API won't process a document that exceeds the maximum size, and will return an invalid document error for it. If an API request has multiple documents, the API will continue processing them if they are within the character limit.
194-
195-
### Rate limits
196-
197-
Your rate limit will vary with your [pricing tier](https://aka.ms/unifiedLanguagePricing).
175+
## Service and data limits
198176

199-
| Tier | Requests per second | Requests per minute |
200-
|---------------|---------------------|---------------------|
201-
| S / Multi-service | 1000 | 1000 |
202-
| S0 / F0 | 100 | 300 |
177+
[!INCLUDE [service limits article](../../includes/service-limits-link.md)]
203178

204179
## See also
205180

‎articles/cognitive-services/language-service/named-entity-recognition/how-to-call.md

Lines changed: 4 additions & 33 deletions
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@ manager: nitinme
88
ms.service: cognitive-services
99
ms.subservice: language-service
1010
ms.topic: how-to
11-
ms.date: 12/10/2021
11+
ms.date: 03/01/2022
1212
ms.author: aahi
1313
ms.custom: language-service-ner, ignite-fall-2021
1414
---
@@ -31,9 +31,7 @@ When you submit documents to be processed, you can specify which of [the support
3131

3232
## Submitting data
3333

34-
Analysis is performed upon receipt of the request. For information on the size and number of requests you can send per minute and second, see the data limits section below.
35-
36-
Using the NER feature synchronously is stateless. No data is stored in your account, and results are returned immediately in the response.
34+
Analysis is performed upon receipt of the request. Using the NER feature synchronously is stateless. No data is stored in your account, and results are returned immediately in the response.
3735

3836
[!INCLUDE [asynchronous-result-availability](../includes/async-result-availability.md)]
3937

@@ -43,36 +41,9 @@ The API will attempt to detect the [defined entity categories](concepts/named-en
4341

4442
When you get results from NER, you can stream the results to an application or save the output to a file on the local system. The API response will include [recognized entities](concepts/named-entity-categories.md), including their categories and sub-categories, and confidence scores.
4543

46-
## Data limits
47-
48-
> [!NOTE]
49-
> * If you need to analyze larger documents than the limit allows, you can break the text into smaller chunks of text before sending them to the API.
50-
> * A document is a single string of text characters.
51-
52-
| Limit | Value |
53-
|------------------------|---------------|
54-
| Maximum size of a single document (synchronous) | 5,120 characters as measured by [StringInfo.LengthInTextElements](/dotnet/api/system.globalization.stringinfo.lengthintextelements). |
55-
| Maximum number of characters per request (asynchronous) | 125K characters across all submitted documents, as measured by [StringInfo.LengthInTextElements](/dotnet/api/system.globalization.stringinfo.lengthintextelements). |
56-
| Maximum size of entire request | 1 MB. |
57-
| Max documents per request | 5 |
58-
59-
If a document exceeds the character limit, the API will behave differently depending on the feature you're using:
60-
61-
* Asynchronous:
62-
* The API will reject the entire request and return a `400 bad request` error if any document within it exceeds the maximum size.
63-
* Synchronous:
64-
* The API won't process a document that exceeds the maximum size, and will return an invalid document error for it. If an API request has multiple documents, the API will continue processing them if they are within the character limit.
65-
66-
Exceeding the maximum number of documents you can send in a single request will generate an HTTP 400 error code.
67-
68-
### Rate limits
69-
70-
Your rate limit will vary with your [pricing tier](https://aka.ms/unifiedLanguagePricing).
44+
## Service and data limits
7145

72-
| Tier | Requests per second | Requests per minute |
73-
|---------------|---------------------|---------------------|
74-
| S / Multi-service | 1000 | 1000 |
75-
| S0 / F0 | 100 | 300 |
46+
[!INCLUDE [service limits article](../includes/service-limits-link.md)]
7647

7748
## Next steps
7849

0 commit comments

Comments
 (0)