v0.0.12

Key changes:

1. Use any model via litellm: Agent(model="litellm/<provider>/<model_name>") e.g. Agent(model="litellm/anthropic/claude-3-5-sonnet-20240620")
2. Enable more complex output types on Agents

What's Changed

• Run CI on all commits, not just ones on main by @rm-openai in #521
• Extract chat completions conversion code into helper by @rm-openai in #522
• Extract chat completions streaming helpers by @rm-openai in #523
• Show repo name/data in docs by @rm-openai in #525
• Litellm integration by @rm-openai in #524
• Docs for LiteLLM integration by @rm-openai in #532
• Docs: Switch to o3 model; exclude translated pages from search by @seratch in #533
• Examples for image inputs by @rm-openai in #553
• Enable non-strict output types by @rm-openai in #539
• Start and finish streaming trace in impl metod by @rm-openai in #540
• Fix visualize graph filename to without extension. by @yuya-haruna in #554
• RFC: automatically use litellm if possible by @rm-openai in #534
• Docs and tests for litellm by @rm-openai in #561
• Pass through organization/project headers to tracing backend, fix speech_group enum by @stevenheidel in #562
• v0.0.12 by @rm-openai in #564

New Contributors

• @yuya-haruna made their first contribution in #554

Full Changelog: v0.0.11...v0.0.12