The answer is quantization and KV cache optimization, as these two techniques directly address the two primary sources of memory consumption during LLM inference: model weights and the autoregressive key-value cache. Quantization reduces the memory footprint by converting model weights and activations from 32-bit floating point to lower precision formats like INT8 or FP16, which cuts storage requirements nearly in half or more. KV cache optimization, on the other hand, reduces memory usage by managing the cached key and value tensors generated during token-by-token decoding, using strategies like pruning, compression, or shared memory to avoid exponential memory growth with longer sequences. On the Oracle Cloud Infrastructure Generative AI Professional 1Z0-1127 exam, this question tests your understanding of practical deployment constraints; a common trap is confusing pruning (a model compression technique) with KV cache optimization, which is specifically about the inference-time cache. Remember the mnemonic "Weights and Cache" — quantization handles the weights, while KV cache optimization handles the cache.
1Z0-1127 Fundamentals of Large Language Models Practice Question
This 1Z0-1127 practice question tests your understanding of fundamentals of large language models. Read the scenario carefully and evaluate each option against the stated constraints before committing to an answer. After answering, compare your reasoning against the explanation and wrong-answer breakdown below. Once you have made your selection, read the full explanation to reinforce the concept and understand why each distractor is designed to mislead on exam day.
Which TWO techniques are commonly used to reduce the memory footprint of LLM inference?
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Quantization
Quantization reduces the memory footprint by lowering the precision of model weights and activations from FP32 to lower bit-widths like INT8 or FP16, which directly decreases the memory required to store and compute with the model. KV cache optimization reduces memory usage by efficiently managing the key-value cache during autoregressive decoding, often through techniques like shared memory, pruning, or compression, which is critical for long-context inference.
Key principle: Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
✓
Quantization
Why this is correct
Reduces memory by using lower precision weights.
Related concept
Read the scenario before looking for a memorised answer.
✗
Increasing batch size
Why it's wrong here
Larger batch size increases memory usage.
✓
KV cache optimization
Why this is correct
Reduces memory for storing key-value tensors during autoregressive generation.
Related concept
Read the scenario before looking for a memorised answer.
✗
Gradient checkpointing
Why it's wrong here
Gradient checkpointing trades compute for memory during training, not inference.
✗
Using full precision (FP32)
Why it's wrong here
Full precision uses more memory than lower precision.
Common exam traps
Common exam trap: answer the scenario, not the keyword
Oracle often tests the distinction between training and inference techniques, so candidates mistakenly apply gradient checkpointing (a training memory saver) to inference, or confuse batch size scaling with memory reduction.
Detailed technical explanation
How to think about this question
Quantization maps floating-point values to a smaller set of discrete levels, often using calibration datasets to minimize accuracy loss, and can be applied post-training (PTQ) or during training (QAT). KV cache optimization can involve techniques like multi-query attention (MQA) or grouped-query attention (GQA), which share key-value heads across query heads to reduce cache size, or use memory-efficient data structures like PagedAttention to manage cache in non-contiguous memory blocks. In real-world deployments, combining 4-bit quantization with KV cache pruning can reduce memory by over 75% for models like LLaMA-2-70B, enabling inference on a single GPU.
KKey Concepts to Remember
Read the scenario before looking for a memorised answer.
Find the constraint that changes the correct option.
Eliminate answers that are true in general but not in this case.
TExam Day Tips
→Watch for words such as best, first, most likely and least administrative effort.
→Review why wrong options are wrong, not only why the correct option is correct.
Key takeaway
Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.
Real-world example
How this comes up in practice
A practitioner preparing for the 1Z0-1127 exam encounters this exact type of scenario on the job. The correct answer here is not the most general option — it is the best answer for the specific constraint described. Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option. Real exam questions reward reading the full scenario before eliminating options, because the constraint defines which answer fits.
What to study next
Got this wrong? Here's your next step.
Identify which exam domain this question belongs to, review the core concept, then practise similar questions from the same domain.
Fundamentals of Large Language Models — This question tests Fundamentals of Large Language Models — Read the scenario before looking for a memorised answer..
What is the correct answer to this question?
The correct answer is: Quantization — Quantization reduces the memory footprint by lowering the precision of model weights and activations from FP32 to lower bit-widths like INT8 or FP16, which directly decreases the memory required to store and compute with the model. KV cache optimization reduces memory usage by efficiently managing the key-value cache during autoregressive decoding, often through techniques like shared memory, pruning, or compression, which is critical for long-context inference.
What should I do if I get this 1Z0-1127 question wrong?
Identify which exam domain this question belongs to, review the core concept, then practise similar questions from the same domain.
What is the key concept behind this question?
Read the scenario before looking for a memorised answer.
About these practice questions
Courseiva creates original exam-style practice questions with explanations and wrong-answer analysis. It does not publish real exam questions, exam dumps, or protected exam content. Learn why practice questions differ from exam dumps →
Share a tip, memory trick, or ask about the reasoning behind this question. Do not post real exam questions, leaked content, braindumps, or copyrighted exam material. Comments are moderated and may be removed without notice.
This 1Z0-1127 practice question is part of Courseiva's free Oracle certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the 1Z0-1127 exam.
Question Discussion
Share a tip, memory trick, or ask about the reasoning behind this question. Do not post real exam questions, leaked content, braindumps, or copyrighted exam material. Comments are moderated and may be removed without notice.
Sign in to join the discussion.