AI Play Book

Foundry Models

1,900+ models from OpenAI, Anthropic, Meta, Microsoft, and partners - deployed through a single Azure endpoint with enterprise SLA, RBAC, content filtering, and data residency.

Your prompts and completions are never used to train models.

Model Catalogue (July 2026)

Provider	Models	Best For
OpenAI	GPT-5.5, GPT-5, GPT-4o, o1, o3-mini, DALL-E 3, Whisper	General purpose, reasoning, multimodal
Anthropic	Claude Opus 5, Claude Sonnet 4	Complex reasoning, long context
Meta	Llama 4, Llama 3.3	Open-weight, self-hosted flexibility
Mistral	Mistral Large, Mistral Small	European data residency
Microsoft	MAI-Thinking-1, MAI-Image-2.5, MAI-Transcribe-2, MAI-Voice-2, Phi-4	First-party, cost-efficient, on-device
DeepSeek	DeepSeek-R1	Reasoning, code
Fireworks AI	Open-model inference	Low latency, high throughput, custom weights
Embeddings	text-embedding-3-large, text-embedding-3-small	RAG, semantic search

Deployment Types

Type	Billing	Latency	Best For
Pay-As-You-Go (PAYG)	Per 1K tokens (input + output)	Variable	Development, testing, variable workloads
Provisioned Throughput (PTU)	Fixed monthly per PTU	Consistent	Production with predictable traffic
Global	PAYG pricing, cheapest region	Higher	Batch processing, latency-insensitive
Managed Compute	Per compute hour	Variable	Regional GPU constraints, custom model hosting
Data Zone (EU/APAC)	PAYG with regional guarantee	Regional	Data residency requirements

PTU is a fixed cost - unused capacity is wasted money. Size to handle P80 traffic and overflow to PAYG for spikes. EU Data Zone and APAC Data Zone pricing effective September 2026.

Getting Started

from openai import AzureOpenAI

client = AzureOpenAI(
    azure_endpoint="https://my-foundry.openai.azure.com/",
    api_key="your-key",
    api_version="2024-12-01-preview"
)

response = client.chat.completions.create(
    model="gpt-5",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Explain RAG in 3 sentences."}
    ]
)

Production (Managed Identity)

from azure.identity import DefaultAzureCredential
from openai import AzureOpenAI

credential = DefaultAzureCredential()
token = credential.get_token("https://cognitiveservices.azure.com/.default")

client = AzureOpenAI(
    azure_endpoint="https://my-foundry.openai.azure.com/",
    azure_ad_token=token.token,
    api_version="2024-12-01-preview"
)

Fine-tuning

Fine-tuning customizes a base model on domain-specific data. Supports GPT-4o, GPT-4o-mini, and open models.

Approach	Use When
Prompt Engineering	Quick iteration, general tasks, no training data
RAG	Reference specific documents, data changes frequently, need citations
Fine-tuning	Domain vocabulary, consistent output format, shorter prompts
Both (RAG + Fine-tune)	Domain writing style AND current document grounding

Frontier Tuning

More than 10x more cost-efficient than GPT-5.5 for domain-specific tasks. Available via the Foundry portal with no hosting fees during experimentation (developer tier).

Cost Optimization

Use max_tokens to cap response length
Semantic caching with AI Search for frequent queries
Smaller models (Phi-4, GPT-4o-mini) for classification/extraction
Route batch workloads to Global deployments
Monitor token usage per deployment with Azure Monitor
Use Managed Compute when you need to route around regional GPU constraints