Skip to content

Commit d02ca68

Browse files
Roopan-MicrosoftRoopan-Microsoft
authored
Merge pull request #873 from chaudhariniraj/dev
chore: Reduce model capacity to 50k
2 parents 577fe45 + 70b96d7 commit d02ca68

9 files changed

Lines changed: 20 additions & 20 deletions

File tree

‎.github/workflows/job-deploy.yml‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -100,7 +100,7 @@ on:
100100
value: ${{ jobs.azure-setup.outputs.QUOTA_FAILED || 'false' }}
101101

102102
env:
103-
GPT_MIN_CAPACITY: 150
103+
GPT_MIN_CAPACITY: 50
104104
IMAGE_MODEL_MIN_CAPACITY: 1
105105
BRANCH_NAME: ${{ github.event.workflow_run.head_branch || github.head_ref || github.ref_name }}
106106
WAF_ENABLED: ${{ inputs.trigger_type == 'workflow_dispatch' && (inputs.waf_enabled || false) || false }}

‎docs/CustomizingAzdParameters.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -14,7 +14,7 @@ By default this template will use the environment name as the prefix to prevent
1414
| `AZURE_ENV_GPT_MODEL_NAME` | string | `gpt-5.1` | Specifies the GPT model name to deploy. |
1515
| `AZURE_ENV_GPT_MODEL_VERSION` | string | `2025-11-13` | Sets the GPT model version. |
1616
| `AZURE_ENV_MODEL_DEPLOYMENT_TYPE` | string | `GlobalStandard` | Defines the model deployment type (allowed: `Standard`, `GlobalStandard`). |
17-
| `AZURE_ENV_GPT_MODEL_CAPACITY` | integer | `150` | Sets the GPT model token capacity (minimum: `10`). |
17+
| `AZURE_ENV_GPT_MODEL_CAPACITY` | integer | `50` | Sets the GPT model token capacity (minimum: `10`). |
1818
| `AZURE_ENV_IMAGE_MODEL_NAME` | string | `gpt-image-1-mini` | Image model to deploy (allowed: `gpt-image-1-mini`, `gpt-image-1.5`, `none`). |
1919
| `AZURE_ENV_IMAGE_MODEL_CAPACITY` | integer | `1` | Sets the image model deployment capacity in RPM (minimum: `1`). |
2020
| `AZURE_ENV_OPENAI_API_VERSION` | string | `2025-01-01-preview` | Specifies the API version for Azure OpenAI service. |

‎docs/DEPLOYMENT.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -139,7 +139,7 @@ When you start the deployment, most parameters will have **default values**, but
139139
| **GPT Model** | Choose from **gpt-4, gpt-4o, gpt-4o-mini, gpt-5.1**. | gpt-5.1 |
140140
| **GPT Model Version** | The version of the selected GPT model. | 2025-11-13 |
141141
| **OpenAI API Version** | The Azure OpenAI API version to use. | 2025-01-01-preview |
142-
| **GPT Model Deployment Capacity** | Configure capacity for **GPT models** (in thousands). | 150k |
142+
| **GPT Model Deployment Capacity** | Configure capacity for **GPT models** (in thousands). | 50k |
143143
| **Image Model** | Choose from **gpt-image-1-mini, gpt-image-1.5** | gpt-image-1-mini |
144144
| **Image Tag** | Docker image tag to deploy. Common values: `latest`, `dev`, `hotfix`. | latest |
145145
| **Existing Log Analytics Workspace** | To reuse an existing Log Analytics Workspace ID. | *(empty)* |
@@ -150,7 +150,7 @@ When you start the deployment, most parameters will have **default values**, but
150150
<details>
151151
<summary><b>[Optional] Quota Recommendations</b></summary>
152152

153-
By default, the **GPT-5.1 model capacity** in deployment is set to **150k tokens**, so we recommend updating the following:
153+
By default, the **GPT-5.1 model capacity** in deployment is set to **50k tokens**, so we recommend updating the following:
154154

155155
> **For GPT-5.1 - increase the capacity post-deployment for optimal performance if required.**
156156

‎docs/QuotaCheck.md‎

Lines changed: 5 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -2,7 +2,7 @@
22

33
Before deploying the Content Generation Solution Accelerator, **ensure sufficient quota availability** for the required models.
44

5-
> **For Global Standard | GPT-5.1 - ensure capacity to at least 150 tokens post-deployment for optimal performance.**
5+
> **For Global Standard | GPT-5.1 - ensure capacity to at least 50 tokens post-deployment for optimal performance.**
66
77
> **For Global Standard | GPT-Image-1-mini - ensure capacity to at least 1 RPM (Requests Per Minute) for image generation.**
88
@@ -15,7 +15,7 @@ az login
1515

1616
### 📌 Default Models & Capacities:
1717
```
18-
gpt-5.1:150,gpt-image-1-mini:1
18+
gpt-5.1:50,gpt-image-1-mini:1
1919
```
2020
**Note:** GPT-5.1 capacity is in tokens, GPT-Image-1-mini capacity is in RPM (Requests Per Minute).
2121
### 📌 Default Regions:
@@ -42,19 +42,19 @@ australiaeast, centralus, eastasia, eastus, eastus2, japaneast, northeurope, sou
4242
```
4343
✔️ Check specific model(s) in default regions:
4444
```
45-
./quota_check_params.sh --models gpt-5.1:150,gpt-image-1-mini:1
45+
./quota_check_params.sh --models gpt-5.1:50,gpt-image-1-mini:1
4646
```
4747
✔️ Check default models in specific region(s):
4848
```
4949
./quota_check_params.sh --regions eastus,swedencentral
5050
```
5151
✔️ Passing Both models and regions:
5252
```
53-
./quota_check_params.sh --models gpt-5.1:150,gpt-image-1-mini:1 --regions eastus,swedencentral
53+
./quota_check_params.sh --models gpt-5.1:50,gpt-image-1-mini:1 --regions eastus,swedencentral
5454
```
5555
✔️ All parameters combined:
5656
```
57-
./quota_check_params.sh --models gpt-5.1:150,gpt-image-1-mini:1 --regions eastus,swedencentral --verbose
57+
./quota_check_params.sh --models gpt-5.1:50,gpt-image-1-mini:1 --regions eastus,swedencentral --verbose
5858
```
5959

6060
### **Sample Output**

‎infra/main.bicep‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -56,7 +56,7 @@ param secondaryLocation string = 'uksouth'
5656
azd: {
5757
type: 'location'
5858
usageName: [
59-
'OpenAI.GlobalStandard.gpt-5.1,150'
59+
'OpenAI.GlobalStandard.gpt-5.1,50'
6060
'OpenAI.GlobalStandard.gpt-image-1-mini,1'
6161
]
6262
}
@@ -92,7 +92,7 @@ param azureOpenaiAPIVersion string = '2025-01-01-preview'
9292

9393
@minValue(10)
9494
@description('Optional. AI model deployment token capacity.')
95-
param gptModelCapacity int = 150
95+
param gptModelCapacity int = 50
9696

9797
@minValue(1)
9898
@description('Optional. Image model deployment capacity (RPM).')

‎infra/main.json‎

Lines changed: 4 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -6,7 +6,7 @@
66
"_generator": {
77
"name": "bicep",
88
"version": "0.44.1.10279",
9-
"templateHash": "11010885094881131352"
9+
"templateHash": "6689866125639874285"
1010
},
1111
"name": "Intelligent Content Generation Accelerator",
1212
"description": "Solution Accelerator for multimodal marketing content generation using Microsoft Agent Framework.\n"
@@ -79,7 +79,7 @@
7979
"azd": {
8080
"type": "location",
8181
"usageName": [
82-
"OpenAI.GlobalStandard.gpt-5.1,150",
82+
"OpenAI.GlobalStandard.gpt-5.1,50",
8383
"OpenAI.GlobalStandard.gpt-image-1-mini,1"
8484
]
8585
},
@@ -134,7 +134,7 @@
134134
},
135135
"gptModelCapacity": {
136136
"type": "int",
137-
"defaultValue": 150,
137+
"defaultValue": 50,
138138
"minValue": 10,
139139
"metadata": {
140140
"description": "Optional. AI model deployment token capacity."
@@ -26158,8 +26158,8 @@
2615826158
},
2615926159
"dependsOn": [
2616026160
"aiFoundryAiServices",
26161-
"[format('avmPrivateDnsZones[{0}]', variables('dnsZoneIndex').openAI)]",
2616226161
"[format('avmPrivateDnsZones[{0}]', variables('dnsZoneIndex').cognitiveServices)]",
26162+
"[format('avmPrivateDnsZones[{0}]', variables('dnsZoneIndex').openAI)]",
2616326163
"virtualNetwork"
2616426164
]
2616526165
},

‎infra/main_custom.bicep‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -57,7 +57,7 @@ param secondaryLocation string = 'uksouth'
5757
azd: {
5858
type: 'location'
5959
usageName: [
60-
'OpenAI.GlobalStandard.gpt-5.1,150'
60+
'OpenAI.GlobalStandard.gpt-5.1,50'
6161
'OpenAI.GlobalStandard.gpt-image-1-mini,1'
6262
]
6363
}
@@ -93,7 +93,7 @@ param azureOpenaiAPIVersion string = '2025-01-01-preview'
9393

9494
@minValue(10)
9595
@description('Optional. AI model deployment token capacity.')
96-
param gptModelCapacity int = 150
96+
param gptModelCapacity int = 50
9797

9898
@minValue(1)
9999
@description('Optional. Image model deployment capacity (RPM).')

‎infra/scripts/quota_check_params.sh‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -37,8 +37,8 @@ log_verbose() {
3737
}
3838

3939
# Default Models and Capacities for Content Generation
40-
# GPT-5.1: 150 tokens, GPT-Image-1-mini: 1 RPM (Requests Per Minute)
41-
DEFAULT_MODEL_CAPACITY="gpt-5.1:150,gpt-image-1-mini:1"
40+
# GPT-5.1: 50 tokens, GPT-Image-1-mini: 1 RPM (Requests Per Minute)
41+
DEFAULT_MODEL_CAPACITY="gpt-5.1:50,gpt-image-1-mini:1"
4242

4343
# Convert the comma-separated string into an array
4444
IFS=',' read -r -a MODEL_CAPACITY_PAIRS <<< "$DEFAULT_MODEL_CAPACITY"

‎scripts/checkquota.sh‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -33,7 +33,7 @@ else
3333
AZURE_ENV_IMAGE_MODEL_NAME="${AZURE_ENV_IMAGE_MODEL_NAME:-gpt-image-1-mini}"
3434
fi
3535

36-
GPT_MIN_CAPACITY="${GPT_MIN_CAPACITY:-150}"
36+
GPT_MIN_CAPACITY="${GPT_MIN_CAPACITY:-50}"
3737
IMAGE_MODEL_MIN_CAPACITY="${IMAGE_MODEL_MIN_CAPACITY:-1}"
3838

3939
# Regions to check

0 commit comments

Comments
 (0)