capacity

Name: capacity
Author: microsoft

1.3kstars215forks

How to install capacity?

microsoft/azure-skills/capacity

$npx -y skills add microsoft/azure-skills --skill capacity

Installs into the current project.

Summary

Discovers available Azure OpenAI model capacity across regions and projects. Analyzes quota limits, compares availability, and recommends optimal deployment locations based on capacity requirements. USE FOR: find capacity, check quota, where can I deploy, capacity discovery, best

Files · 1

View on GitHub

SKILL.md

1	# Capacity Discovery
2
3	Finds available Azure OpenAI model capacity across all accessible regions and projects. Recommends the best deployment location based on capacity requirements.
4
5	## Quick Reference
6
7	\| Property \| Description \|
8	\|----------\|-------------\|
9	\| Purpose \| Find where you can deploy a model with sufficient capacity \|
10	\| Scope \| All regions and projects the user has access to \|
11	\| Output \| Ranked table of regions/projects with available capacity \|
12	\| Action \| Read-only analysis — does NOT deploy. Hands off to preset or customize \|
13	\| Authentication \| Azure CLI (`az login`) \|
14
15	## When to Use This Skill
16
17	- ✅ User asks "where can I deploy gpt-4o?"
18	- ✅ User specifies a capacity target: "find a region with 10K TPM for gpt-4o"
19	- ✅ User wants to compare availability: "which regions have gpt-4o available?"
20	- ✅ User got a quota error and needs to find an alternative location
21	- ✅ User asks "best region and project for deploying model X"
22
23	After discovery → hand off to [preset](../preset/SKILL.md) or [customize](../customize/SKILL.md) for actual deployment.
24
25	## Scripts
26
27	Pre-built scripts handle the complex REST API calls and data processing. Use these instead of constructing commands manually.
28
29	\| Script \| Purpose \| Usage \|
30	\|--------\|---------\|-------\|
31	\| `scripts/discover_and_rank.ps1` \| Full discovery: capacity + projects + ranking \| Primary script for capacity discovery \|
32	\| `scripts/discover_and_rank.sh` \| Same as above (bash) \| Primary script for capacity discovery \|
33	\| `scripts/query_capacity.ps1` \| Raw capacity query (no project matching) \| Quick capacity check or version listing \|
34	\| `scripts/query_capacity.sh` \| Same as above (bash) \| Quick capacity check or version listing \|
35
36	## Workflow
37
38	### Phase 1: Validate Prerequisites
39
40	```bash
41	az account show --query "{Subscription:name, SubscriptionId:id}" --output table
42	```
43
44	### Phase 2: Identify Model and Version
45
46	Extract model name from user prompt. If version is unknown, query available versions:
47
48	```powershell
49	.\scripts\query_capacity.ps1 -ModelName <model-name>
50	```
51	```bash
52	./scripts/query_capacity.sh <model-name>
53	```
54
55	This lists available versions. Use the latest version unless user specifies otherwise.
56
57	### Phase 3: Run Discovery
58
59	Run the full discovery script with model name, version, and minimum capacity target:
60
61	```powershell
62	.\scripts\discover_and_rank.ps1 -ModelName <model-name> -ModelVersion <version> -MinCapacity <target>
63	```
64	```bash
65	./scripts/discover_and_rank.sh <model-name> <version> <min-capacity>
66	```
67
68	> 💡 The script automatically queries capacity across ALL regions, cross-references with the user's existing projects, and outputs a ranked table sorted by: meets target → project count → available capacity.
69
70	### Phase 3.5: Validate Subscription Quota
71
72	After discovery identifies candidate regions, validate that the user's subscription actually has available quota in each region. Model capacity (from Phase 3) shows what the platform can support, but subscription quota limits what this specific user can deploy.
73
74	```powershell
75	# For each candidate region from discovery results:
76	$usageData = az cognitiveservices usage list --location <region> --subscription $SUBSCRIPTION_ID -o json 2>$null \| ConvertFrom-Json
77
78	# Check quota for each SKU the model supports
79	# Quota names follow pattern: OpenAI.<SKU>.<model-name>
80	$usageEntry = $usageData \| Where-Object { $_.name.value -eq "OpenAI.<SKU>.<model-name>" }
81
82	if ($usageEntry) {
83	$quotaAvailable = $usageEntry.limit - $usageEntry.currentValue
84	} else {
85	$quotaAvailable = 0 # No quota allocated
86	}
87	```
88	```bash
89	# For each candidate region from discovery results:
90	usage_json=$(az cognitiveservices usage list --location <region> --subscription "$SUBSCRIPTION_ID" -o json 2>/dev/null)
91
92	# Extract quota for specific SKU+model
93	quota_available=$(echo "$usage_json" \| jq -r --arg name "OpenAI.<SKU>.<model-name>" \
94	'.[] \| select(.name.value == $name) \| .limit - .currentValue')
95	```
96
97	Annotate discovery results:
98
99	Add a "Quota Available" column to the ranked output from Phase 3:
100
101	\| Region \| Available Capacity \| Meets Target \| Projects \| Quota Available \|
102	\|--------\|-------------------\|--------------\|----------\|-----------------\|
103	\| eastus2 \| 120K TPM \| ✅ \| 3 \| ✅ 80K \|
104	\| westus3 \| 90K TPM \| ✅ \| 1 \| ❌ 0 (at limit) \|
105	\| swedencentral \| 100K TP

People who liked this also viewed

6 picks

Type

capacity

1.3kstars215forks

View on GitHub

How to install capacity?

microsoft/azure-skills/capacity

$npx -y skills add microsoft/azure-skills --skill capacity

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Capacity Discovery
2
3	Finds available Azure OpenAI model capacity across all accessible regions and projects. Recommends the best deployment location based on capacity requirements.
4
5	## Quick Reference
6
7	\| Property \| Description \|
8	\|----------\|-------------\|
9	\| Purpose \| Find where you can deploy a model with sufficient capacity \|
10	\| Scope \| All regions and projects the user has access to \|
11	\| Output \| Ranked table of regions/projects with available capacity \|
12	\| Action \| Read-only analysis — does NOT deploy. Hands off to preset or customize \|
13	\| Authentication \| Azure CLI (`az login`) \|
14
15	## When to Use This Skill
16
17	- ✅ User asks "where can I deploy gpt-4o?"
18	- ✅ User specifies a capacity target: "find a region with 10K TPM for gpt-4o"
19	- ✅ User wants to compare availability: "which regions have gpt-4o available?"
20	- ✅ User got a quota error and needs to find an alternative location
21	- ✅ User asks "best region and project for deploying model X"
22
23	After discovery → hand off to [preset](../preset/SKILL.md) or [customize](../customize/SKILL.md) for actual deployment.
24
25	## Scripts
26
27	Pre-built scripts handle the complex REST API calls and data processing. Use these instead of constructing commands manually.
28
29	\| Script \| Purpose \| Usage \|
30	\|--------\|---------\|-------\|
31	\| `scripts/discover_and_rank.ps1` \| Full discovery: capacity + projects + ranking \| Primary script for capacity discovery \|
32	\| `scripts/discover_and_rank.sh` \| Same as above (bash) \| Primary script for capacity discovery \|
33	\| `scripts/query_capacity.ps1` \| Raw capacity query (no project matching) \| Quick capacity check or version listing \|
34	\| `scripts/query_capacity.sh` \| Same as above (bash) \| Quick capacity check or version listing \|
35
36	## Workflow
37
38	### Phase 1: Validate Prerequisites
39
40	```bash
41	az account show --query "{Subscription:name, SubscriptionId:id}" --output table
42	```
43
44	### Phase 2: Identify Model and Version
45
46	Extract model name from user prompt. If version is unknown, query available versions:
47
48	```powershell
49	.\scripts\query_capacity.ps1 -ModelName <model-name>
50	```
51	```bash
52	./scripts/query_capacity.sh <model-name>
53	```
54
55	This lists available versions. Use the latest version unless user specifies otherwise.
56
57	### Phase 3: Run Discovery
58
59	Run the full discovery script with model name, version, and minimum capacity target:
60
61	```powershell
62	.\scripts\discover_and_rank.ps1 -ModelName <model-name> -ModelVersion <version> -MinCapacity <target>
63	```
64	```bash
65	./scripts/discover_and_rank.sh <model-name> <version> <min-capacity>
66	```
67
68	> 💡 The script automatically queries capacity across ALL regions, cross-references with the user's existing projects, and outputs a ranked table sorted by: meets target → project count → available capacity.
69
70	### Phase 3.5: Validate Subscription Quota
71
72	After discovery identifies candidate regions, validate that the user's subscription actually has available quota in each region. Model capacity (from Phase 3) shows what the platform can support, but subscription quota limits what this specific user can deploy.
73
74	```powershell
75	# For each candidate region from discovery results:
76	$usageData = az cognitiveservices usage list --location <region> --subscription $SUBSCRIPTION_ID -o json 2>$null \| ConvertFrom-Json
77
78	# Check quota for each SKU the model supports
79	# Quota names follow pattern: OpenAI.<SKU>.<model-name>
80	$usageEntry = $usageData \| Where-Object { $_.name.value -eq "OpenAI.<SKU>.<model-name>" }
81
82	if ($usageEntry) {
83	$quotaAvailable = $usageEntry.limit - $usageEntry.currentValue
84	} else {
85	$quotaAvailable = 0 # No quota allocated
86	}
87	```
88	```bash
89	# For each candidate region from discovery results:
90	usage_json=$(az cognitiveservices usage list --location <region> --subscription "$SUBSCRIPTION_ID" -o json 2>/dev/null)
91
92	# Extract quota for specific SKU+model
93	quota_available=$(echo "$usage_json" \| jq -r --arg name "OpenAI.<SKU>.<model-name>" \
94	'.[] \| select(.name.value == $name) \| .limit - .currentValue')
95	```
96
97	Annotate discovery results:
98
99	Add a "Quota Available" column to the ranked output from Phase 3:
100
101	\| Region \| Available Capacity \| Meets Target \| Projects \| Quota Available \|
102	\|--------\|-------------------\|--------------\|----------\|-----------------\|
103	\| eastus2 \| 120K TPM \| ✅ \| 3 \| ✅ 80K \|
104	\| westus3 \| 90K TPM \| ✅ \| 1 \| ❌ 0 (at limit) \|
105	\| swedencentral \| 100K TP

People who liked this also viewed

6 picks

Type