Why model abstraction matters in LangChain - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a simple model abstraction using LangChain.

LangChain

from langchain.llms import OpenAI

llm = [1](temperature=0.7)

print(llm("Say hello"))

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

AOpenAI

BPromptTemplate

CBaseLLM

DChatOpenAI

Attempts:

3 left

💡 Hint

Common Mistakes

Using ChatOpenAI instead of OpenAI here

Using PromptTemplate which is for prompts, not models

The OpenAI class is used to create a language model instance in LangChain for text generation.

2fill in blank

medium

Complete the code to use model abstraction to switch between models easily.

LangChain

def get_model(model_name):
    if model_name == "openai":
        return [1](temperature=0)
    elif model_name == "chat":
        return ChatOpenAI(temperature=0)

model = get_model("openai")
print(model("Hello!"))

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

AOpenAI

BBaseLLM

CPromptTemplate

DChatOpenAI

Attempts:

3 left

💡 Hint

Common Mistakes

Returning BaseLLM which is abstract

Returning PromptTemplate which is not a model

Using OpenAI here allows the function to return the correct model instance based on the name, demonstrating abstraction.

3fill in blank

hard

Fix the error in the code by choosing the correct method to generate text from the model abstraction.

LangChain

llm = OpenAI(temperature=0.5)

response = llm.[1](["What is AI?"])
print(response.generations[0][0].text)

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

Arun

Bgenerate

Ccall

Dexecute

Attempts:

3 left

💡 Hint

Common Mistakes

Using call which does not exist on the LLM instance

Using run or execute which do not exist here

The generate method is the correct way to invoke the model instance to generate text in LangChain. It takes a list of prompts and returns an LLMResult.

4fill in blank

hard

Fill both blanks to create a flexible prompt and generate a response using model abstraction.

LangChain

from langchain.prompts import [1]
from langchain.llms import OpenAI

prompt = [2](template="Say something about {topic}.")
llm = OpenAI(temperature=0.3)

text = prompt.format(topic="LangChain")
response = llm(text)
print(response)

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

APromptTemplate

BChatPromptTemplate

CBasePrompt

DSimplePrompt

Attempts:

3 left

💡 Hint

Common Mistakes

Using ChatPromptTemplate which is for chat models

Using BasePrompt which is abstract

PromptTemplate is used to create flexible prompts with placeholders, which can then be formatted with actual values.

5fill in blank

hard

Fill all three blanks to build a chain that uses model abstraction and prompt templates.

LangChain

from langchain.chains import LLMChain
from langchain.llms import [1]
from langchain.prompts import [2]

llm = [1](temperature=0.2)
prompt = [2](template="Explain {concept} simply.")
chain = LLMChain(llm=llm, prompt=prompt)

result = chain.run(concept="model abstraction")
print(result)

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

AOpenAI

BChatOpenAI

CPromptTemplate

DBaseLLM

Attempts:

3 left

💡 Hint

Common Mistakes

Using ChatOpenAI with PromptTemplate which may mismatch

Using BaseLLM which is abstract and cannot be instantiated

Using OpenAI for the model and PromptTemplate for the prompt creates a chain that abstracts the model and prompt logic.