SLM High-Fidelity Few-Shot Template (Llama 3 / Phi-3)

Compact, XML-structured few-shot prompt template designed for extracting 98%+ formatting accuracy from 8B parameter small language models.

#SLM#Llama 3#Phi-3#Few-Shot#JSON

Interactive Prompt Customizer

Fill in the variable parameters below to generate a tailored Llama 3 8B & Phi-3 prompt.

Input Text to Extract ({{user_input_text}})

System Prompt (Llama 3 8B & Phi-3)

<system>
You are an expert JSON Extractor. Extract entity data from the input text strictly matching the JSON format.
Rules:
1. Output ONLY valid JSON.
2. Do NOT add markdown code block tags or extra text.
</system>

Tailored User Prompt Template

<example_1>
<input>John Doe is a 34 year old Software Engineer living in Austin, TX.</input>
<output>{"name":"John Doe","age":34,"role":"Software Engineer","city":"Austin","state":"TX"}</output>
</example_1>

<live_task>
<input>Alex Rivera, 41, Principal Cloud Architect at Google in San Francisco, CA.</input>
<output>

Ready to execute in Llama 3 8B & Phi-3

Featured AI Platform

Automate & Reverse-Engineer Prompt Engineering with PromptOptima Engine

Verified Example Output

Output generated when running this customized prompt template in Llama 3 8B & Phi-3:

{"name":"Alex Rivera","age":41,"role":"Principal Cloud Architect","company":"Google","city":"San Francisco","state":"CA"}

PromptOptima SaaS Integration

Want to auto-tune this prompt for minimum token cost?

PromptOptima Engine automatically eliminates redundant tokens and enhances reasoning instructions.

1-Click Reverse Engineering 35% Token Cost Reduction

Try PromptOptima Engine

Frequently Asked Questions

Why use XML tags for SLMs?

XML tags provide explicit token boundary markers that prevent SLM instruction drift.