部署微調的模型以進行測試

若要部署候選模型，請選取要部署的微調模型，然後選取 [ 部署]。

[部署模型] 對話方塊隨即開啟。 在對話框中，輸入您的 [部署名稱 ]，然後從 [部署類型] 下拉式清單中選取 [ 開發人員 ]。 選取 [建立] 以開始部署自定義模型。

你可以在 Microsoft Foundry 入口網站的 部署 面板中監控新部署的進度。

import json
import os
import requests

token = os.getenv("<TOKEN>") 
subscription = "<YOUR_SUBSCRIPTION_ID>"  
resource_group = "<YOUR_RESOURCE_GROUP_NAME>"
resource_name = "<YOUR_AZURE_OPENAI_RESOURCE_NAME>"
model_deployment_name = "gpt41-mini-candidate-01" # custom deployment name that you will use to reference the model when making inference calls.

deploy_params = {'api-version': "2025-07-01-preview"} 
deploy_headers = {'Authorization': 'Bearer {}'.format(token), 'Content-Type': 'application/json'}

deploy_data = {
    "sku": {"name": "developertier", "capacity": 50},
    "properties": {
        "model": {
            "format": "OpenAI",
            "name": <"fine_tuned_model">, #retrieve this value from the previous call, it will look like gpt41-mini-candidate-01.ft-b044a9d3cf9c4228b5d393567f693b83
            "version": "1"
        }
    }
}
deploy_data = json.dumps(deploy_data)

request_url = f'https://management.azure.com/subscriptions/{subscription}/resourceGroups/{resource_group}/providers/Microsoft.CognitiveServices/accounts/{resource_name}/deployments/{model_deployment_name}'

print('Creating a new deployment...')

r = requests.put(request_url, params=deploy_params, headers=deploy_headers, data=deploy_data)

print(r)
print(r.reason)
print(r.json())

下列範例示範如何使用 REST API 來為您的自定義模型建立模型部署。 REST API 會產生自訂模型部署的名稱。

curl -X POST "https://management.azure.com/subscriptions/<SUBSCRIPTION>/resourceGroups/<RESOURCE_GROUP>/providers/Microsoft.CognitiveServices/accounts/<RESOURCE_NAME>/deployments/<MODEL_DEPLOYMENT_NAME>?api-version=2025-07-01-preview" \
  -H "Authorization: Bearer <TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{
    "sku": {"name": "developertier", "capacity": 50},
    "properties": {
        "model": {
            "format": "OpenAI",
            "name": "<FINE_TUNED_MODEL>",
            "version": "1"
        }
    }
}'

使用 Azure CLI 部署模型

下列範例示範如何使用 Azure CLI 來部署自訂模型。 使用 Azure CLI，您必須指定自訂模型部署的名稱。 如需如何使用 Azure CLI 部署自訂模型的詳細資訊，請參閱 az cognitiveservices account deployment。

若要在主控台視窗中執行此 Azure CLI 命令，您必須使用自訂模型的對應值取代下列<預留位置>：

az cognitiveservices account deployment create 
    --resource-group <YOUR_RESOURCE_GROUP>
    --name <YOUR_RESOURCE_NAME>  
    --deployment-name <YOUR_DEPLOYMENT_NAME>
    --model-name <YOUR_FINE_TUNED_MODEL_ID>
    --model-version "1" 
    --model-format OpenAI 
    --sku-capacity "50" 
    --sku-name "Developer"

部署自訂模型後，您就能如同使用其他部署模型般使用該模型。 您可以使用 Foundry 入口網站中的 [遊樂場] 對新部署進行實驗。 您可以繼續對自訂模型使用相同的參數，例如 temperature 和 max_tokens，如同對其他部署模型一樣。

您也可以使用 Evaluations 服務，針對已部署的模型候選專案和其他模型版本建立和執行模型評估。

import os
from openai import OpenAI

client = OpenAI(
  base_url = "https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/", 
  api_key=os.getenv("AZURE_OPENAI_API_KEY"),  
)

response = client.chat.completions.create(
    model="gpt41-mini-candidate-01", # model = "Custom deployment name you chose for your fine-tuning model"
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Does Azure OpenAI support customer managed keys?"},
        {"role": "assistant", "content": "Yes, customer managed keys are supported by Azure OpenAI."},
        {"role": "user", "content": "Do other Foundry Tools support this too?"}
    ]
)

print(response.choices[0].message.content)

curl $AZURE_OPENAI_ENDPOINT/openai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "api-key: $AZURE_OPENAI_API_KEY" \
  -d '"model": "YOUR_MODEL_DEPLOYMENT_NAME", {"messages":[{"role": "system", "content": "You are a helpful assistant."},{"role": "user", "content": "Does Azure OpenAI support customer managed keys?"},{"role": "assistant", "content": "Yes, customer managed keys are supported by Azure OpenAI."},{"role": "user", "content": "Do other Foundry Tools support this too?"}]}'