Skip to content

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

0.1.0 - 2025-01-XX

Added

  • Initial release of ullm
  • Support for OpenAI, Anthropic, Groq, and AWS Bedrock providers
  • Chat completion API (completion() and acompletion())
  • OpenAI Responses API (responses() and aresponses())
  • Streaming support for all providers
  • Tool calling/function calling support
  • Structured output with Pydantic models
  • Exponential backoff retry logic using tenacity
  • Comprehensive test suite
  • CI/CD with GitHub Actions
  • Full API compatibility with litellm core features
  • Examples and documentation

Features

  • 100x smaller memory footprint than litellm (~2MB vs ~200MB)
  • 24x faster import time
  • Modern Python tooling (uv, ruff, mypy)
  • Type hints throughout
  • Async/await support
  • Minimal dependencies (httpx, pydantic, tenacity)