Azure OpenAI Limits

A Python package that provides static token limits for Azure OpenAI models, including context and output limits for various model versions.

Installation

pip install azure-openai-limits

Usage

Python API

from azure_openai_limits import get_limits, context_limit, output_limit

# Get limits for a model (uses default version)
limits = get_limits("gpt-4o")
print(f"Context: {limits.context}, Output: {limits.output}")

# Get limits for a specific version
limits = get_limits("gpt-4o", "2024-05-13")
print(f"Context: {limits.context}, Output: {limits.output}")

# Try the new O3 Pro model
limits = get_limits("o3-pro")
print(f"O3 Pro - Context: {limits.context}, Output: {limits.output}")

# Check Codex Mini limits
limits = get_limits("codex-mini")
print(f"Codex Mini - Context: {limits.context}, Output: {limits.output}")

# Get just context or output limits
context = context_limit("gpt-4o")
output = output_limit("gpt-4o")

# Check if a model/version exists
from azure_openai_limits import model_exists
if model_exists("gpt-4o", "2024-08-06"):
    print("Model version exists!")

# List all available models
from azure_openai_limits import list_models
models = list_models()
print("Available models:", models)

# List all versions for a model
from azure_openai_limits import list_versions
versions = list_versions("gpt-4o")
print("Available versions:", versions)

Command Line Interface

# List all available models and their versions
azure-openai-limits list

# Show limits for a specific model (default version)
azure-openai-limits show gpt-4o

# Show limits for a specific model version
azure-openai-limits show gpt-4o --version 2024-05-13

# Check out the new O3 Pro model
azure-openai-limits show o3-pro

Supported Models

The package includes limits for the following Azure OpenAI models:

• Codex: codex-mini
• GPT-3.5: gpt-35-turbo, gpt-35-turbo-instruct
• GPT-4: gpt-4, gpt-4-32k
• GPT-4.1: gpt-4.1, gpt-4.1-mini, gpt-4.1-nano
• GPT-4o: gpt-4o, gpt-4o-mini
• GPT-4o Audio: gpt-4o-audio-preview, gpt-4o-mini-audio-preview
• GPT-4o Realtime: gpt-4o-realtime-preview, gpt-4o-mini-realtime-preview
• GPT-5: gpt-5, gpt-5-mini, gpt-5-nano, gpt-5-chat, gpt-5-pro, gpt-5-codex
• GPT-5.1: gpt-5.1, gpt-5.1-chat, gpt-5.1-codex, gpt-5.1-codex-mini, gpt-5.1-codex-max
• GPT-5.2: gpt-5.2, gpt-5.2-chat
• GPT-5.3: gpt-5.3-chat, gpt-5.3-codex
• GPT-5.4: gpt-5.4, gpt-5.4-pro
• GPT OSS: gpt-oss-20b, gpt-oss-120b
• O1 Models: o1, o1-mini, o1-preview
• O3 Models: o3, o3-mini, o3-pro
• O4 Models: o4-mini
• Model Router: model-router

Each model may have multiple versions with different limits. Use azure-openai-limits list to see all available models and their versions.

Data Source

The model limits data in this package is sourced from the official Microsoft documentation: Azure AI Foundry OpenAI Models

Last Updated: March 16, 2026

API Reference

Classes

Limits

A dataclass representing token limits for a model.

Attributes:

• context: int - Maximum context tokens (input + conversation history)
• output: int - Maximum output tokens that can be generated
• total_max: int - Property returning context + output

Functions

get_limits(model: str, version: str | None = None) -> Limits

Get context and output limits for a specific model.

context_limit(model: str, version: str | None = None) -> int

Get just the context limit for a model.

output_limit(model: str, version: str | None = None) -> int

Get just the output limit for a model.

model_exists(model: str, version: str | None = None) -> bool

Check if a model (and optionally version) exists.

list_models() -> list[str]

Get a sorted list of all available model names.

list_versions(model: str) -> list[str]

Get a sorted list of all versions for a specific model.

all_models() -> dict[str, dict[str, Limits]]

Get the complete model data structure.

Error Handling

The package raises appropriate exceptions:

• KeyError: When a model or version is not found
• ValueError: When invalid input is provided (e.g., empty model name)

All error messages include helpful information about available models/versions.

Development

To set up for development:

git clone <repository-url>
cd azure-openai-limits
pip install -e ".[dev]"

Run tests:

pytest

Run linting:

ruff check src/
mypy src/

License

MIT License