added

Author: brown9804

0_Azure/3_AzureAI/9_AzureOpenAI/demos/4_PTUs_TPM.md

@@ -9,17 +9,27 @@ Last updated: 2025-07-17
 
 ----------
 
-> Provisioned Throughput Units (PTUs) <br/>
-> Tokens Per Minute (TPM)
-
-| **PTUs** | **Calls per Minute** | **Tokens in Prompt** | **Tokens in Response** | **Tokens per Minute (TPM)** |
-|----------|----------------------|----------------------|------------------------|-----------------------------|
-| 1        | 10                   | 50                   | 100                    | 1,500                       |
-| 2        | 20                   | 50                   | 100                    | 3,000                       |
-| 5        | 50                   | 50                   | 100                    | 7,500                       |
-| 10       | 100                  | 50                   | 100                    | 15,000                      |
-| 20       | 200                  | 50                   | 100                    | 30,000                      |
-| 50       | 500                  | 50                   | 100                    | 75,000                      |
+`Provisioned Throughput Units (PTUs)` <br/>
+`Tokens Per Minute (TPM)`
+
+<details>
+<summary><b>List of References </b> (Click to expand)</summary>
+
+- [How much throughput per PTU you get for each model](https://learn.microsoft.com/en-us/azure/ai-foundry/openai/how-to/provisioned-throughput-onboarding)
+- [Understanding costs associated with provisioned throughput units (PTU)](https://learn.microsoft.com/en-us/azure/ai-foundry/openai/how-to/provisioned-throughput-onboarding#azure-reservations-for-azure-ai-foundry-provisioned-throughput)
+- [Deployment types for Azure AI Foundry Models](https://learn.microsoft.com/en-us/azure/ai-foundry/openai/how-to/deployment-types#global-provisioned)
+- [Region availability for provisioned throughput capability](https://learn.microsoft.com/en-us/azure/ai-foundry/openai/concepts/provisioned-throughput?tabs=global-ptum#region-availability-for-provisioned-throughput-capability)
+- [Model summary table and region availability](https://learn.microsoft.com/en-us/azure/ai-foundry/openai/concepts/models?tabs=global-ptum%2Cstandard-chat-completions#model-summary-table-and-region-availability)
+- [Fine-tuning models](https://learn.microsoft.com/en-us/azure/ai-foundry/openai/concepts/models?tabs=global-ptum%2Cstandard-chat-completions#fine-tuning-models) - input/output Max
+- [Azure OpenAI in Azure AI Foundry Models quotas and limits](https://learn.microsoft.com/en-us/azure/ai-foundry/openai/quotas-limits?context=%2Fazure%2Fai-foundry%2Fcontext%2Fcontext&tabs=REST)
+
+</details>
+
+<div align="center">
+  <img width="700" alt="image" src="https://github.com/user-attachments/assets/0741d4b2-d70e-4b5e-a6cf-9c399483e598" style="border: 2px solid #4CAF50; border-radius: 5px; padding: 5px;"/>
+</div>
+
+From [Microsoft official documentation](https://learn.microsoft.com/en-us/azure/ai-foundry/openai/how-to/provisioned-throughput-onboarding#model-independent-quota)
 
 > Explanation:
 
@@ -29,35 +39,29 @@ Last updated: 2025-07-17
 - **Tokens in Response**: The number of tokens in the model's response for each call.
 - **Tokens per Minute (TPM)**: The total number of tokens processed per minute, calculated as:
 
+1. **Tokens per Minute**: Calculate the total tokens per minute:
+
 $$
 \text{TPM} = \text{Calls per Minute} \times (\text{Tokens in Prompt} + \text{Tokens in Response})
 $$
 
-> Example Calculation:
-For 50 PTUs:
-
-1. **Calls per Minute**: Calculate the number of calls per minute:
-
-$$
-\text{Calls per Minute} = \text{PTUs} \times \text{Calls per PTU per Minute}
-$$
+2. **Provisioned Throughput Units**: 
 
 $$
-\text{Calls per Minute} = 50 \times 10 = 500
+\text{PTUs} = \frac{\text{TPM}}{\text{Tokens per PTU per Minute}}
 $$
 
-2. **Tokens per Minute**: Calculate the total tokens per minute:
+Where:
+- **TPM** = Total tokens you want to process per minute
+- **Tokens per PTU per Minute** = Depends on the model (e.g., 3,000 tokens/min for GPT-4.1 or GPT-4.1 Mini)
 
-$$
-\text{TPM} = \text{Calls per Minute} \times (\text{Tokens in Prompt} + \text{Tokens in Response})
-$$
+> E.g
+> If you want to process **150,000 tokens per minute** using GPT-4.1:
 
 $$
-\text{TPM} = 500 \times (50 + 100) = 500 \times 150 = 75,000
+\text{PTUs} = \frac{150{,}000}{3{,}000} = 50
 $$
 
-This means with 50 PTUs, you can process 75,000 tokens per minute.
-
 ## Provisioned Capacity Calculator
 
 > Improve accuracy of your estimate by adding multiple workloads to your PTU calculation. Each workload will be calculated and displayed as well as the aggregate total if both are running at the same time to your deployment.