Skip to content

Commit ef430c0

Browse files
Merge pull request #251233 from mrbullwinkle/mrb_09_12_2023_model_update
[Azure AI Series] Update Models page
2 parents 87c4972 + 3ed6452 commit ef430c0

File tree

1 file changed

+81
-23
lines changed
  • articles/ai-services/openai/concepts

1 file changed

+81
-23
lines changed

articles/ai-services/openai/concepts/models.md

Lines changed: 81 additions & 23 deletions
Original file line numberDiff line numberDiff line change
@@ -5,7 +5,7 @@ description: Learn about the different model capabilities that are available wit
55
ms.service: cognitive-services
66
ms.subservice: openai
77
ms.topic: conceptual
8-
ms.date: 09/05/2023
8+
ms.date: 09/12/2023
99
ms.custom: event-tier1-build-2022, references_regions, build-2023, build-2023-dataai
1010
manager: nitinme
1111
author: mrbullwinkle #ChrisHMSFT
@@ -20,8 +20,8 @@ Azure OpenAI Service is powered by a diverse set of models with different capabi
2020

2121
| Models | Description |
2222
|--|--|
23-
| [GPT-4](#gpt-4) | A set of models that improve on GPT-3.5 and can understand as well as generate natural language and code. |
24-
| [GPT-3.5](#gpt-35) | A set of models that improve on GPT-3 and can understand as well as generate natural language and code. |
23+
| [GPT-4](#gpt-4) | A set of models that improve on GPT-3.5 and can understand and generate natural language and code. |
24+
| [GPT-3.5](#gpt-35) | A set of models that improve on GPT-3 and can understand and generate natural language and code. |
2525
| [Embeddings](#embeddings-models) | A set of models that can convert text into numerical vector form to facilitate text similarity. |
2626
| [DALL-E](#dall-e-models-preview) (Preview) | A series of models in preview that can generate original images from natural language. |
2727

@@ -52,14 +52,9 @@ Like GPT-4, use the Chat Completions API to use GPT-3.5 Turbo. To learn more abo
5252
> [!IMPORTANT]
5353
> We strongly recommend using `text-embedding-ada-002 (Version 2)`. This model/version provides parity with OpenAI's `text-embedding-ada-002`. To learn more about the improvements offered by this model, please refer to [OpenAI's blog post](https://openai.com/blog/new-and-improved-embedding-model). Even if you are currently using Version 1 you should migrate to Version 2 to take advantage of the latest weights/updated token limit. Version 1 and Version 2 are not interchangeable, so document embedding and document search must be done using the same version of the model.
5454
55+
The previous embeddings models have been consolidated into the following new replacement model:
5556

56-
Currently, we offer three families of Embeddings models for different functionalities:
57-
The following list indicates the length of the numerical vector returned by the service, based on model capability:
58-
59-
| Base Model | Model(s) | Dimensions |
60-
|---|---|---|
61-
| Ada | models ending in -001 (Version 1) | 1024 |
62-
| Ada | text-embedding-ada-002 (Version 2) | 1536 |
57+
`text-embedding-ada-002`
6358

6459
## DALL-E (Preview)
6560

@@ -99,18 +94,17 @@ GPT-3.5 Turbo is used with the Chat Completion API. GPT-3.5 Turbo (0301) can als
9994

10095
<sup>1</sup> Version `0301` of gpt-35-turbo will be retired no earlier than July 5, 2024. See [model updates](#model-updates) for model upgrade behavior.
10196

102-
10397
### Embeddings models
10498

10599
These models can only be used with Embedding API requests.
106100

107101
> [!NOTE]
108102
> We strongly recommend using `text-embedding-ada-002 (Version 2)`. This model/version provides parity with OpenAI's `text-embedding-ada-002`. To learn more about the improvements offered by this model, please refer to [OpenAI's blog post](https://openai.com/blog/new-and-improved-embedding-model). Even if you are currently using Version 1 you should migrate to Version 2 to take advantage of the latest weights/updated token limit. Version 1 and Version 2 are not interchangeable, so document embedding and document search must be done using the same version of the model.
109103
110-
| Model ID | Base model Regions | Fine-Tuning Regions | Max Request (tokens) | Training Data (up to) |
104+
| Model ID | Base model Regions | Fine-Tuning Regions | Max Request (tokens) | Training Data (up to) | Output dimensions |
111105
| --- | --- | --- | --- | --- |
112-
| text-embedding-ada-002 (version 2) | Canada East, East US, France Central, Japan East, North Central US, South Central US, Switzerland North, UK South, West Europe | N/A |8,191 | Sep 2021 |
113-
| text-embedding-ada-002 (version 1) | East US, South Central US, West Europe | N/A |2,046 | Sep 2021 |
106+
| text-embedding-ada-002 (version 2) | Canada East, East US, East US2, France Central, Japan East, North Central US, South Central US, Switzerland North, UK South, West Europe | N/A |8,191 | Sep 2021 | 1536 |
107+
| text-embedding-ada-002 (version 1) | East US, South Central US, West Europe | N/A |2,046 | Sep 2021 | 1536 |
114108

115109
### DALL-E models (Preview)
116110

@@ -132,13 +126,13 @@ Azure OpenAI now supports automatic updates for select model deployments. On mod
132126

133127
### Auto update to default
134128

135-
When **Auto-update to default** is selected your model deployment will be automatically updated within two weeks of a new version being released.
129+
When **Auto-update to default** is selected your model deployment will be automatically updated within two weeks of a change in the default version.
136130

137-
If you are still in the early testing phases for completion and chat completion based models, we recommend deploying models with **auto-update to default** set whenever it is available.
131+
If you are still in the early testing phases for inference models, we recommend deploying models with **auto-update to default** set whenever it is available.
138132

139133
### Specific model version
140134

141-
As your use of Azure OpenAI evolves, and you start to build and integrate with applications you will likely want to manually control model updates so that you can first test and validate that model performance is remaining consistent for your use case prior to upgrade.
135+
As your use of Azure OpenAI evolves, and you start to build and integrate with applications you may want to manually control model updates so that you can first test and validate that model performance is remaining consistent for your use case prior to upgrade.
142136

143137
When you select a specific model version for a deployment this version will remain selected until you either choose to manually update yourself, or once you reach the retirement date for the model. When the retirement date is reached the model will auto-upgrade to the default version at the time of retirement.
144138

@@ -156,6 +150,76 @@ To view deprecation/expiration dates for all available models in a given region
156150

157151
:::image type="content" source="../media/models/column-options.png" alt-text="Screenshot of the models UI of Azure OpenAI Studio." lightbox="../media/models/column-options.png":::
158152

153+
### Model deployment upgrade configuration
154+
155+
There are three distinct model deployment upgrade options which are configurable via REST API:
156+
157+
| Name | Description |
158+
|------|--------|
159+
| `OnceNewDefaultVersionAvailable` | Once a new version is designated as the default, the model deployment will auto-upgrade to the default version within two weeks of that designation change being made. |
160+
`OnceCurrentVersionExpired` | Once the retirement date is reached the model deployment will auto-upgrade to the current default version. |
161+
`NoAutoUpgrade` | The model deployment will never auto-upgrade. Once the retirement date is reached the model deployment will stop working. You will need to update your code referencing that deployment to point to a non-expired model deployment. |
162+
163+
To query the current model deployment settings including the deployment upgrade configuration for a given resource use [`Deployments List`](/rest/api/cognitiveservices/accountmanagement/deployments/list?tabs=HTTP#code-try-0)
164+
165+
```http
166+
GET https://management.azure.com/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.CognitiveServices/accounts/{accountName}/deployments?api-version=2023-05-01
167+
```
168+
169+
**Path parameters**
170+
171+
| Parameter | Type | Required? | Description |
172+
|--|--|--|--|
173+
| ```acountname``` | string | Required | The name of your Azure OpenAI Resource. |
174+
| ```resourceGroupName``` | string | Required | The name of the associated resource group for this model deployment. |
175+
| ```subscriptionId``` | string | Required | Subscription ID for the associated subscription. |
176+
| ```api-version``` | string | Required |The API version to use for this operation. This follows the YYYY-MM-DD format. |
177+
178+
**Supported versions**
179+
180+
- `2023-05-01` [Swagger spec](https://github.com/Azure/azure-rest-api-specs/blob/1e71ad94aeb8843559d59d863c895770560d7c93/specification/cognitiveservices/resource-manager/Microsoft.CognitiveServices/stable/2023-05-01/cognitiveservices.json)
181+
182+
### Example response
183+
184+
```json
185+
{
186+
"id": "/subscriptions/{Subcription-GUID}/resourceGroups/{Resource-Group-Name}/providers/Microsoft.CognitiveServices/accounts/{Resource-Name}/deployments/text-davinci-003",
187+
"type": "Microsoft.CognitiveServices/accounts/deployments",
188+
"name": "text-davinci-003",
189+
"sku": {
190+
"name": "Standard",
191+
"capacity": 60
192+
},
193+
"properties": {
194+
"model": {
195+
"format": "OpenAI",
196+
"name": "text-davinci-003",
197+
"version": "1"
198+
},
199+
"versionUpgradeOption": "OnceNewDefaultVersionAvailable",
200+
"capabilities": {
201+
"completion": "true",
202+
"search": "true"
203+
},
204+
"raiPolicyName": "Microsoft.Default",
205+
"provisioningState": "Succeeded",
206+
"rateLimits": [
207+
{
208+
"key": "request",
209+
"renewalPeriod": 10,
210+
"count": 60
211+
},
212+
{
213+
"key": "token",
214+
"renewalPeriod": 60,
215+
"count": 60000
216+
}
217+
]
218+
}
219+
```
220+
221+
You can then take the settings from this list to construct an update model REST API call as described below if you want to modify the deployment upgrade configuration.
222+
159223
### Update & deploy models via the API
160224

161225
```http
@@ -245,12 +309,6 @@ curl -X PUT https://management.azure.com/subscriptions/00000000-0000-0000-0000-0
245309
}
246310
```
247311

248-
249-
250-
251-
252-
253-
254312
## Next steps
255313

256314
- [Learn more about Azure OpenAI](../overview.md)

0 commit comments

Comments
 (0)