Set up a Python development environment on Windows with PyCharm

Set up authentication

API calls require an AccessKey pair from a RAM user. For more information, see Create an AccessKey.

Generate and run the SDK sample code

1. Navigate to the SDK Center.

2. Select the service (such as ECS) and the API operation (such as DescribeInstanceTypeFamilies).

3. On the Parameters tab, fill in the required parameters. For this example, set RegionId to cn-qingdao and Generation to ecs-5.

4. In the rightmost column, click the SDK Sample Code tab, select Python, and click Download Project.

   In the SDK sample area on the right, set the SDK generation to V2.0 Recommended and the language to Python.

5. Decompress the downloaded package and open the project in PyCharm.

6. After you run the code, a statusCode value of 200 in the response indicates that the API call was successful, and the response body contains the list of ECS instance type families. The following is part of the response:

   : '1rHpYSAeV5ewkgoDCes4x8A7'}, 'statusCode': 200, 'body': {'InstanceTypeFamilies': {'InstanceTypeFamily': [{'Generation': 'ecs-5', 'InstanceTypeFamilyId': 'ecs.g6'}, {'Generation': 'ecs-5', 'InstanceTypeFamilyId': 'ecs.ebmg

For more information, see Use the Alibaba Cloud Python SDK in an IDE.

Set up authentication

Activate Model Studio and get an API key. For more information, see Account setup.

Install the SDK and run the sample code

1. In the PyCharm terminal, install the OpenAI Python SDK:

   pip install -U openai

   Downloading jiter-0.8.2-cp312-cp312-win_amd64.whl (204 kB)
      204.7/204.7 kB 654.9 kB/s eta 0:00:00
   Downloading pydantic-2.10.6-py3-none-any.whl (431 kB)
      431.7/431.7 kB 642.5 kB/s eta 0:00:00
   Downloading pydantic_core-2.27.2-cp312-cp312-win_amd64.whl (2.0 MB)
      2.0/2.0 MB 715.4 kB/s eta 0:00:00
   Downloading sniffio-1.3.1-py3-none-any.whl (10 kB)
   Downloading tqdm-4.67.1-py3-none-any.whl (78 kB)
      78.5/78.5 kB 875.6 kB/s eta 0:00:00
   Downloading typing_extensions-4.12.2-py3-none-any.whl (37 kB)
   Downloading annotated_types-0.7.0-py3-none-any.whl (13 kB)
   Downloading h11-0.14.0-py3-none-any.whl (58 kB)
      58.3/58.3 kB 762.2 kB/s eta 0:00:00
   Installing collected packages: typing-extensions, tqdm, sniffio, jiter, h11, distro, annotated-types, pydantic-core, httpcore, anyio, pydantic, httpx, openai
   Successfully installed annotated-types-0.7.0 anyio-4.8.0 distro-1.9.0 h11-0.14.0 httpcore-1.0.7 httpx-0.28.1 jiter-0.8.2 openai-1.65.2 pydantic-2.10.6 pydantic-core-2.27.2 sniffio-1.3.1 tqdm-4.67.1 typing-extensions-4.12.2

   A Successfully installed ... openai-x.x.x message confirms the installation.

2. Create a new Python file and add the following code:

   import os
   from openai import OpenAI
   
   try:
       client = OpenAI(
           # Reads the API key from the DASHSCOPE_API_KEY environment variable.
           api_key=os.getenv("DASHSCOPE_API_KEY"),
           base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
       )
   
       completion = client.chat.completions.create(
           model="qwen-plus",
           messages=[
               {'role': 'system', 'content': 'You are a helpful assistant.'},
               {'role': 'user', 'content': 'Who are you?'}
           ]
       )
       print(completion.choices[0].message.content)
   except Exception as e:
       print(f"An error occurred: {e}")

3. Run the script. You should see a response from the Qwen model.

   D:\official_site\Demo\venv\Scripts\python.exe D:\official_site\Demo\venv\ecsram\TongYi.py
   I am a large language model from Alibaba Cloud. My name is Qwen.

For more information, see the OpenAI Python SDK.

Set up authentication

Activate Model Studio and get an API key. For more information, see Account setup.

Install the SDK and run the sample code

1. In the PyCharm terminal, install the OpenAI Python SDK:

   pip install -U openai

   Downloading jiter-0.8.2-cp312-cp312-win_amd64.whl (204 kB)
      204.7/204.7 kB 654.9 kB/s eta 0:00:00
   Downloading pydantic-2.10.6-py3-none-any.whl (431 kB)
      431.7/431.7 kB 642.5 kB/s eta 0:00:00
   Downloading pydantic_core-2.27.2-cp312-cp312-win_amd64.whl (2.0 MB)
      2.0/2.0 MB 715.4 kB/s eta 0:00:00
   Downloading sniffio-1.3.1-py3-none-any.whl (10 kB)
   Downloading tqdm-4.67.1-py3-none-any.whl (78 kB)
      78.5/78.5 kB 875.6 kB/s eta 0:00:00
   Downloading typing_extensions-4.12.2-py3-none-any.whl (37 kB)
   Downloading annotated_types-0.7.0-py3-none-any.whl (13 kB)
   Downloading h11-0.14.0-py3-none-any.whl (58 kB)
      58.3/58.3 kB 762.2 kB/s eta 0:00:00
   Installing collected packages: typing-extensions, tqdm, sniffio, jiter, h11, distro, annotated-types, pydantic-core, httpcore, anyio, pydantic, httpx, openai
   Successfully installed annotated-types-0.7.0 anyio-4.8.0 distro-1.9.0 h11-0.14.0 httpcore-1.0.7 httpx-0.28.1 jiter-0.8.2 openai-1.65.2 pydantic-2.10.6 pydantic-core-2.27.2 sniffio-1.3.1 tqdm-4.67.1 typing-extensions-4.12.2

   A Successfully installed ... openai-x.x.x message confirms the installation.

2. Create a new Python file and add the following code:

   import os
   from openai import OpenAI
   
   client = OpenAI(
       # os.getenv retrieves the API Key from an environment variable. If you have not configured the environment variable,
       # replace the next line with your Alibaba Cloud Bailian API Key, for example: api_key="sk-xxx",
       api_key=os.getenv("DASHSCOPE_API_KEY"),
       base_url="https://dashscope.aliyuncs.com/compatible-mode/v1"
   )
   
   completion = client.chat.completions.create(
       model="deepseek-r1",  # The model `deepseek-r1` is used here as an example. You can replace it with a different model name as needed.
       messages=[
           {'role': 'user', 'content': 'Which is larger, 9.9 or 9.11?'}
       ]
   )
   
   # Print the reasoning process via the reasoning_content field
   print("Reasoning Process:")
   print(completion.choices[0].message.reasoning_content)
   
   # Print the final answer via the content field
   print("Final Answer:")
   print(completion.choices[0].message.content)
   

3. After you run the preceding code, the following sample output is returned.

   Reasoning Process:
   The user asks which is larger, 9.9 or 9.11. The question looks simple, but it needs careful analysis because the answer depends on how the numbers are represented. The first thing to settle is whether the user means decimals or another form of number, such as a version number or a date. For a question phrased this way in mathematics, decimal comparison is intended.
   
   Next, the structure of the two numbers has to be clear. 9.9 is normally read as 9 plus 0.9, that is, 9.90, whereas 9.11 is 9 plus 0.11. The integer part of both numbers is 9, so the comparison comes down to the fractional parts: 0.90 against 0.11. Because 0.90 is greater than 0.11, 9.9 is greater than 9.11. A common misunderstanding arises when the numbers are treated as version numbers, as in software versions 9.9 and 9.11.
   
   The user might also read 9.11 as 9 and 11/100, that is 0.11, and 9.9 as 9 and 0.9. In that case 0.9 is clearly greater than 0.11, so 9.9 is larger. People sometimes compare only the tenths digit, 1 in 9.11 against 9 in 9.9, and happen to reach the same conclusion. The correct method is to align the number of decimal places, rewrite 9.9 as 9.90, and then compare it with 9.11.
   
   In summary, the answer differs depending on whether the values are decimals or version numbers, but by mathematical convention, as decimals, 9.9 is larger. If the user means version numbers, the answer is the opposite, so the context could be confirmed. For a general mathematics question, the answer is that 9.9 is larger.
   Final Answer:
   
   When 9.9 and 9.11 are compared as **decimals in mathematics**:
   - Pad 9.9 to 9.90, whose fractional part is 0.90.
   - The fractional part of 9.11 is 0.11.
   - Because **0.90 > 0.11**, **9.9 > 9.11**.
   
   If **version numbers** (such as software versions) are involved:
   - 9.11 may indicate the 11th update of version 9, whereas 9.9 is the 9th update of version 9. In that case, **9.11 > 9.9**.
   
   **By default, the values are compared as decimals, so 9.9 is larger**.

Getting started with DeepSeek.