You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: articles/ai-services/openai/concepts/models.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -66,7 +66,7 @@ See [model versions](../concepts/model-versions.md) to learn about how Azure Ope
66
66
67
67
| Model ID | Description | Max Request (tokens) | Training Data (up to) |
68
68
| --- | :--- |:--- |:---: |
69
-
|`gpt-4o` (2024-08-06) <br> **GPT-4o (Omni)**|**Latest large GA model** <br> - Structured outputs<br> - Text, image processing <br> - JSON Mode <br> - parallel function calling <br> - Enhanced accuracy and responsiveness <br> - Parity with English text and coding tasks compared to GPT-4 Turbo with Vision <br> - Superior performance in non-English languages and in vision tasks |Input: 128,000 <br> Output: 4,096| Oct 2023 |
69
+
|`gpt-4o` (2024-08-06) <br> **GPT-4o (Omni)**|**Latest large GA model** <br> - Structured outputs<br> - Text, image processing <br> - JSON Mode <br> - parallel function calling <br> - Enhanced accuracy and responsiveness <br> - Parity with English text and coding tasks compared to GPT-4 Turbo with Vision <br> - Superior performance in non-English languages and in vision tasks |Input: 128,000 <br> Output: 16,384 | Oct 2023 |
70
70
|`gpt-4o-mini` (2024-07-18) <br> **GPT-4o mini**|**Latest small GA model** <br> - Fast, inexpensive, capable model ideal for replacing GPT-3.5 Turbo series models. <br> - Text, image processing <br>- JSON Mode <br> - parallel function calling | Input: 128,000 <br> Output: 16,384 | Oct 2023 |
71
71
|`gpt-4o` (2024-05-13) <br> **GPT-4o (Omni)**| Text, image processing <br> - JSON Mode <br> - parallel function calling <br> - Enhanced accuracy and responsiveness <br> - Parity with English text and coding tasks compared to GPT-4 Turbo with Vision <br> - Superior performance in non-English languages and in vision tasks |Input: 128,000 <br> Output: 4,096| Oct 2023 |
72
72
|`gpt-4` (turbo-2024-04-09) <br>**GPT-4 Turbo with Vision**|**New GA model** <br> - Replacement for all previous GPT-4 preview models (`vision-preview`, `1106-Preview`, `0125-Preview`). <br> - [**Feature availability**](#gpt-4o-and-gpt-4-turbo) is currently different depending on method of input, and deployment type. | Input: 128,000 <br> Output: 4,096 | Dec 2023 |
Copy file name to clipboardExpand all lines: articles/search/search-limits-quotas-capacity.md
+34-55Lines changed: 34 additions & 55 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -8,7 +8,7 @@ author: HeidiSteen
8
8
ms.author: heidist
9
9
ms.service: cognitive-search
10
10
ms.topic: conceptual
11
-
ms.date: 06/13/2024
11
+
ms.date: 09/04/2024
12
12
ms.custom:
13
13
- references_regions
14
14
- build-2024
@@ -63,11 +63,14 @@ You might find some variation in maximum limits if your service happens to be pr
63
63
64
64
## Document limits
65
65
66
-
You can have approximately 24 billion documents per index on Basic, S1, S2, S3, L1, and L2 search services. For S3 HD, the limit is 2 billion documents per index. Each instance of a complex collection counts as a separate document in terms of these limits.
66
+
Maximum number of documents per index are:
67
67
68
-
### Document size limits per API call
68
+
+ 24 billion on Basic, S1, S2, S3, L1, and L2 search services.
69
+
+ 2 billion on S3 HD.
69
70
70
-
The maximum document size when calling an Index API is approximately 16 megabytes.
71
+
Each instance of a complex collection counts as a separate document in terms of these limits.
72
+
73
+
Maximum document size when calling an Index API is approximately 16 megabytes.
71
74
72
75
Document size is actually a limit on the size of the Index API request body. Since you can pass a batch of multiple documents to the Index API at once, the size limit realistically depends on how many documents are in the batch. For a batch with a single document, the maximum document size is 16 MB of JSON.
73
76
@@ -79,73 +82,49 @@ When you index documents with vector fields, Azure AI Search constructs internal
79
82
80
83
The service enforces a vector index size quota **for every partition** in your search service. Each extra partition increases the available vector index size quota. This quota is a hard limit to ensure your service remains healthy, which means that further indexing attempts once the limit is exceeded results in failure. You can resume indexing once you free up available quota by either deleting some vector documents or by scaling up in partitions.
81
84
82
-
The table describes the vector index size quota per partition across the service tiers. For context, it includes:
83
-
84
-
+[Partition storage limits](#service-limits) for each tier, repeated here for context.
85
-
+ Amount of each partition (in GB) available for vector indexes (created when you add vector fields to an index).
86
-
+ Approximate number of embeddings (floating point values) per partition.
85
+
Vector limits vary by service creation date and tier.
87
86
88
-
Use the [GET Service Statistics](/rest/api/searchservice/get-service-statistics) to retrieve your vector index size quota or review the **Indexes** page or **Usage** tab in the Azure portal.
87
+
+ To check the age of your search service or learn more about vector indexes, see [Vector index size and staying under limits](vector-search-index-size.md).
89
88
90
-
Vector limits vary by service creation date and tier. To check the age of your search service and learn more about vector indexes, see [Vector index size and staying under limits](vector-search-index-size.md).
89
+
+ To view the vector quota in effect for your search service, use [GET Service Statistics](/rest/api/searchservice/get-service-statistics), or check the **Properties** and **Usage** tabs for your search service in the Azure portal.
91
90
92
-
###Vector limits on services created after May 17, 2024
91
+
#### Storage quota (GB)
93
92
94
-
The highest vector limits are available on search services created after May 17, 2024 in a [supported region](#service-limits).
93
+
This table repeats [partition storage limits](#service-limits) for context. The table shows the progression of storage quota increases in GB over time. Vector quota is per partition, so the increase in vector quota is bound to the increase in per-partition storage for each tier. Higher capacity partitions were brought online starting in April 2024.
<sup>3</sup> Higher capacity storage for Basic, S1, S2, S3 in the following regions. **Americas**: Brazil South, Canada Central, Canada East, East US, East US 2, Central US, North Central US, South Central US, West US, West US 2, West US 3, West Central US. **Europe**: France Central. Italy North, North Europe, Norway East, Poland Central, Switzerland North, Sweden Central, UK South, UK West. **Middle East**: UAE North. **Africa**: South Africa North. **Asia Pacific**: Australia East, Australia Southeast, Central India, Jio India West, East Asia, Southeast Asia, Japan East, Japan West, Korea Central, Korea South.
117
107
118
-
Notice that L1 and L2 limits are unchanged in the April 3 rollout.
108
+
<sup>4</sup> Higher capacity storage for more tiers and more regions. **Europe**: Germany North, Germany West Central, Switzerland West. **Azure Government**: Texas, Arizona, Virginia. **Africa**: South Africa North. **Asia Pacific**: China North 3, China East 3.
119
109
120
-
### Vector limits on services created between July 1, 2023 and April 3, 2024
110
+
####Vector quota per partition (GB)
121
111
122
-
The following limits applied to new services created between July 1 and April 3, 2024, except for the following regions, which have the original limits from before July 1, 2023:
112
+
This table shows the progression of vector quota increases in GB over time. The quota is per partition, so if you scale a new Standard (S1) service to 6 partitions, total vector quota is 35 multiplied by 6.
123
113
124
-
+ Germany West Central
125
-
+ West India
126
-
+ Qatar Central
114
+
| Service creation date |Basic | S1| S2 | S3 | L1 | L2 |
<sup>2</sup> Vector limits during the later preview period. Three regions didn't have the higher limits: Germany West Central, West India, Qatar Central.
138
124
139
-
### Vector limits on services created before July 1, 2023
125
+
<sup>3</sup> Higher vector quota based on the larger partitions for supported tiers and regions.
Copy file name to clipboardExpand all lines: articles/search/search-region-support.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -21,7 +21,7 @@ This article identifies the cloud regions in which Azure AI Search is available.
21
21
22
22
| Feature | Availability |
23
23
|---------|--------------|
24
-
|[Extra capacity](search-limits-quotas-capacity.md#service-limits)| Higher capacity partitions became available in selected regions starting in April 2024 with a second wave following in May 2024. Currently, there are just a few regions that *don't* offer higher capacity partitions. If you're using an older search service, create a new search service to benefit from more capacity at the same billing rate. To check existing capacity, [find your search service](https://portal.azure.com/#blade/HubsExtension/BrowseResourceBlade/resourceType/Microsoft.Search%2FsearchServices) and select the **Properties** tab in the middle of the Overview page. To check search service age, follow [these instructions](vector-search-index-size.md#how-to-check-service-creation-date). Regional support for extra capacity is noted in the footnotes of this article.|
24
+
|[Extra capacity](search-limits-quotas-capacity.md#service-limits)| Higher capacity partitions became available in selected regions starting in April 2024 with a second wave following in May 2024. If you're using an older search service, create a new search service to benefit from more capacity at the same billing rate. To check existing capacity, [find your search service](https://portal.azure.com/#blade/HubsExtension/BrowseResourceBlade/resourceType/Microsoft.Search%2FsearchServices) and select the **Properties** tab in the middle of the Overview page. To check search service age, follow [these instructions](vector-search-index-size.md#how-to-check-service-creation-date). Currently, there are just a few regions that *don't* offer higher capacity partitions. Regional support for extra capacity is noted in the footnotes of this article.|
25
25
|[Availability zones](search-reliability.md#availability-zone-support)| Divides a region's data centers into distinct physical location groups, providing high-availability within the same geo. Regional support is noted in this article. |
26
26
|[Azure AI enrichment](cognitive-search-concept-intro.md)| Refers to skills that make internal calls to Azure AI for enrichment and transformation during indexing. Integration requires that Azure AI Search coexists with an [Azure AI multi-service account](/azure/ai-services/multi-service-resource) in the same physical region. Regional support is noted in this article. |
27
27
|[Azure OpenAI integration](vector-search-integrated-vectorization.md)| Refers to skills and vectorizers that make internal calls to deployed embedding and chat models on Azure OpenAI. Check [Azure OpenAI model region availability](/azure/ai-services/openai/concepts/models#model-summary-table-and-region-availability) for the most current list of regions for each embedding and chat model. Specific Azure OpenAI models are in fewer regions, so be sure to check for joint regional availability before installing.|
Copy file name to clipboardExpand all lines: articles/search/search-reliability.md
+5-4Lines changed: 5 additions & 4 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -6,7 +6,7 @@ author: mattgotteiner
6
6
ms.author: magottei
7
7
ms.service: cognitive-search
8
8
ms.topic: conceptual
9
-
ms.date: 01/02/2024
9
+
ms.date: 09/04/2024
10
10
ms.custom:
11
11
- subject-reliability
12
12
- references_regions
@@ -55,10 +55,9 @@ Availability zones are used when you add two or more replicas to your search ser
55
55
56
56
### Supported regions
57
57
58
-
Support for availability zones depends on infrastructure and storage. Currently, two zones that were announced in October 2023 have insufficient storage and don't provide an availability zone for Azure AI Search:
58
+
Support for availability zones depends on infrastructure and storage. Currently, the following zone has insufficient storage and doesn't provide an availability zone for Azure AI Search:
59
59
60
-
+ Israel Central
61
-
+ Italy North
60
+
+ Japan West
62
61
63
62
Otherwise, availability zones for Azure AI Search are supported in the following regions:
64
63
@@ -75,6 +74,8 @@ Otherwise, availability zones for Azure AI Search are supported in the following
Copy file name to clipboardExpand all lines: articles/search/search-what-is-azure-search.md
+4-4Lines changed: 4 additions & 4 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -41,13 +41,13 @@ On the search service itself, the two primary workloads are *indexing* and *quer
41
41
42
42
+[**Indexing**](search-what-is-an-index.md) is an intake process that loads content into your search service and makes it searchable. Internally, inbound text is processed into tokens and stored in inverted indexes, and inbound vectors are stored in vector indexes. The document format that Azure AI Search can index is JSON. You can upload JSON documents that you've assembled, or use an indexer to retrieve and serialize your data into JSON.
43
43
44
-
[Applied AI](cognitive-search-concept-intro.md) through a [skillset](cognitive-search-working-with-skillsets.md) extends indexing with image and language models. If you have images or large unstructured text in source document, you can attach skills that perform OCR, analyze and escribe images, infer structure, translate text and more.
44
+
[Applied AI](cognitive-search-concept-intro.md) through a [skillset](cognitive-search-working-with-skillsets.md) extends indexing with image and language models. If you have images or large unstructured text in source document, you can attach skills that perform OCR, analyze and describe images, infer structure, translate text and more. Output is text that can be serialized into JSON and ingested into a search index.
45
45
46
-
Skillsets can also perform [data chunking and vectorization during indexing](vector-search-integrated-vectorization.md). Skills that attach to Azure OpenAI, the model catalog in Azure AI Studio, or custom skills that attach to any external chunking and embedding model can be used during indexing to create vector data.
46
+
Skillsets can also perform [data chunking and vectorization during indexing](vector-search-integrated-vectorization.md). Skills that attach to Azure OpenAI, the model catalog in Azure AI Studio, or custom skills that attach to any external chunking and embedding model can be used during indexing to create vector data. Output is chunked vector content that can be ingested into a search index.
47
47
48
48
+[**Querying**](search-query-overview.md) can happen once an index is populated with searchable content, when your client app sends query requests to a search service and handles responses. All query execution is over a search index that you control.
49
49
50
-
[Semantic ranking](semantic-search-overview.md) is an extension of query execution. It adds secondary ranking, using language understanding to reevalute a result set, promoting the most semantically relevant results to the top.
50
+
[Semantic ranking](semantic-search-overview.md) is an extension of query execution. It adds secondary ranking, using language understanding to reevaluate a result set, promoting the most semantically relevant results to the top.
51
51
52
52
[Integrated vectorization](vector-search-integrated-vectorization.md) is also an extension of query execution. If you have vector fields in your search index, you can submit raw vector queries or text that's vectorized at query time.
53
53
@@ -99,7 +99,7 @@ Or, try solution accelerators:
99
99
100
100
+[**Chat with your data** solution accelerator](https://github.com/Azure-Samples/chat-with-your-data-solution-accelerator) helps you create a custom RAG solution over your content.
101
101
102
-
+[**Conversational Knowledge Mining** solution accelerator](https://github.com/microsoft/Customer-Service-Conversational-Insights-with-Azure-OpenAI-Services), helps you create an interactive solution to extract actionable insights from post-contact center transcripts.
102
+
+[**Conversational Knowledge Mining** solution accelerator](https://github.com/microsoft/Customer-Service-Conversational-Insights-with-Azure-OpenAI-Services) helps you create an interactive solution to extract actionable insights from post-contact center transcripts.
103
103
104
104
+[**Build your own copilot** solution accelerator](https://github.com/microsoft/Build-your-own-copilot-Solution-Accelerator), leverages Azure OpenAI Service, Azure AI Search and Microsoft Fabric, to create custom copilot solutions.
0 commit comments