Workshop Samples - Quick Reference Card

Last Updated: October 8, 2025

---

🚀 Quick Start

# 1. Ensure Foundry Local is running
foundry service status
foundry model run phi-4-mini

# 2. Install dependencies
pip install -r Workshop/requirements.txt

# 3. Run a sample
cd Workshop/samples
python -m session01.chat_bootstrap "What is edge AI?"

---

📂 Sample Overview

Session	Sample	Purpose	Time
01	chat_bootstrap.py	Basic chat + streaming	~30s
02	rag_pipeline.py	RAG with embeddings	~45s
02	rag_eval_ragas.py	RAG evaluation	~60s
03	benchmark_oss_models.py	Model benchmarking	~2m
04	model_compare.py	SLM vs LLM	~45s
05	agents_orchestrator.py	Multi-agent system	~60s
06	models_router.py	Intent routing	~45s
06	models_pipeline.py	Multi-step pipeline	~60s

---

🛠️ Environment Variables

Essential

# Choose model
set FOUNDRY_LOCAL_ALIAS=phi-4-mini

# Override endpoint (optional)
set FOUNDRY_LOCAL_ENDPOINT=http://localhost:8000

# Show token usage
set SHOW_USAGE=1

Session-Specific

# Session 02: RAG
set RAG_QUESTION="What is local inference?"
set EMBED_MODEL=sentence-transformers/all-MiniLM-L6-v2

# Session 03: Benchmarking
set BENCH_MODELS=phi-4-mini,qwen2.5-0.5b
set BENCH_ROUNDS=3
set BENCH_STREAM=1

# Session 04: Comparison
set SLM_ALIAS=phi-4-mini
set LLM_ALIAS=qwen2.5-7b

# Session 05: Agents
set AGENT_MODEL_PRIMARY=phi-4-mini
set AGENT_QUESTION="Why use edge AI?"

# Session 06: Pipeline
set PIPELINE_TASK="Your task here"

---

✅ Validation & Testing

# Validate syntax and imports
python scripts/validate_samples.py

# Validate specific session
python scripts/validate_samples.py --session 01

# Run smoke tests
python scripts/test_samples.py --quick

# Verbose testing
python scripts/test_samples.py --verbose

---

🐛 Troubleshooting

Connection Error

# Check Foundry Local
foundry service status

# Start if needed
foundry service start
foundry model run phi-4-mini

Import Error

# Install missing dependencies
pip install sentence-transformers ragas datasets

# Or install all
pip install -r Workshop/requirements.txt

Model Not Found

# List available models
foundry model ls

# Download model
foundry model download phi-4-mini

Slow Performance

# Use smaller model
set FOUNDRY_LOCAL_ALIAS=qwen2.5-0.5b

# Reduce benchmark rounds
set BENCH_ROUNDS=1

---

📖 Common Patterns

Basic Chat

from workshop_utils import chat_once

text, usage = chat_once(
    'phi-4-mini',
    messages=[{"role": "user", "content": "Hello"}],
    max_tokens=100,
    temperature=0.7
)

Get Client

from workshop_utils import get_client

manager, client, model_id = get_client(
    alias='phi-4-mini',
    endpoint=None  # Auto-detect
)

Error Handling

try:
    manager, client, model_id = get_client(alias)
except Exception as e:
    print(f"[ERROR] Failed: {e}")
    print("[INFO] Check: foundry service status")
    sys.exit(1)

Streaming

stream = client.chat.completions.create(
    model=model_id,
    messages=messages,
    stream=True
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

---

📊 Model Selection

Model	Size	Best For	Speed
qwen2.5-0.5b	0.5B	Fast classification	⚡⚡⚡
qwen2.5-coder-0.5b	0.5B	Quick code gen	⚡⚡⚡
gemma-2-2b	2B	Creative writing	⚡⚡
phi-3.5-mini	3.5B	Code, refactoring	⚡⚡
phi-4-mini	4B	General, summary	⚡⚡
qwen2.5-7b	7B	Complex reasoning	⚡

---

🔗 Resources

• SDK Docs: https://github.com/microsoft/Foundry-Local/tree/main/sdk/python
• Quick Ref: Workshop/FOUNDRY_SDK_QUICKREF.md

---

💡 Tips

1. Cache clients: workshop_utils caches for you
2. Use smaller models: Start with qwen2.5-0.5b for testing
3. Enable usage stats: Set SHOW_USAGE=1 to track tokens
4. Batch processing: Process multiple prompts sequentially
5. Lower max_tokens: Reduces latency for quick responses

---

🎯 Sample Workflows

Test Everything

python scripts/validate_samples.py
python scripts/test_samples.py --quick

Benchmark Models

cd samples
set BENCH_MODELS=phi-4-mini,qwen2.5-0.5b
set BENCH_ROUNDS=3
python -m session03.benchmark_oss_models

RAG Pipeline

cd samples
set RAG_QUESTION="What is RAG?"
python -m session02.rag_pipeline

Multi-Agent System

cd samples
set AGENT_QUESTION="Why edge AI for healthcare?"
python -m session05.agents_orchestrator

---

Quick Help: Run any sample with --help from the samples directory or check the docstring:

python -c "import session01.chat_bootstrap; help(session01.chat_bootstrap)"

---

All samples updated October 2025 with Foundry Local SDK best practices ✨