LangChain vs direct API calls - Interactive Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the LangChain OpenAI class.

LangChain

from langchain.llms import [1]

Drag options to blanks, or click blank then click option'

AOpenAIClient

BChatOpenAI

CLangOpenAI

DOpenAI

Attempts:

3 left

💡 Hint

Common Mistakes

Using ChatOpenAI which is for chat models, not direct OpenAI calls.

Trying to import a non-existent class like OpenAIClient.

The LangChain library provides the OpenAI class to interact with OpenAI models easily.

2fill in blank

medium

Complete the code to create a LangChain OpenAI instance with temperature 0.7.

LangChain

llm = OpenAI(temperature=[1])

Drag options to blanks, or click blank then click option'

A0.7

B1.0

C0

DNone

Attempts:

3 left

💡 Hint

Common Mistakes

Using 0 which makes output deterministic.

Using None which is invalid for temperature.

The temperature controls randomness. Setting it to 0.7 adds some creativity.

3fill in blank

hard

Fix the error in the direct OpenAI API call to generate text.

LangChain

response = client.completions.create(model="text-davinci-003", [1]="Hello, world!", max_tokens=50)

Drag options to blanks, or click blank then click option'

Amessage

Binput_text

Cprompt

Dtext

Attempts:

3 left

💡 Hint

Common Mistakes

Using input_text which is for embeddings.create.

Using message which is for chat.completions.create.

The direct OpenAI Python client expects the prompt parameter for the input text in completions.create.

4fill in blank

hard

Fill both blanks to create a LangChain prompt template and generate text.

LangChain

from langchain.prompts import [1]
prompt = [2](template="Say hello to {name}!")
result = llm(prompt.format(name="Alice"))

Drag options to blanks, or click blank then click option'

APromptTemplate

BPrompt

CTemplatePrompt

DPromptFormatter

Attempts:

3 left

💡 Hint

Common Mistakes

Using Prompt which is not a class in LangChain prompts module.

Using TemplatePrompt which does not exist.

The PromptTemplate class is used to create templates in LangChain.

5fill in blank

hard

Fill all three blanks to create a direct OpenAI chat call with messages and get the reply.

LangChain

messages = [{"role": [1], "content": [2]]
response = client.chat.completions.create(model="gpt-4o-mini", messages=messages)
reply = response.choices[0].[3]

Drag options to blanks, or click blank then click option'

A"user"

B"Hello, how are you?"

Cmessage.content

D"system"

Attempts:

3 left

💡 Hint

Common Mistakes

Using "system" role for user messages.

Accessing reply as text instead of message.content.

The chat message role is "user", content is the text string, and the reply is accessed via message.content.