v0.0.44

What's Changed

• Migrate OpenAI models away from max_tokens to max_completion_tokens by @barapa in #1206
• Add return docstring to the function description by @Kludex in #1207
• Add model_request_parameters attribute (containing tool definitions) to chat spans by @alexmojaki in #1177
• Use get_running_loop instead of get_event_loop by @Kludex in #1204
• Bump openai to 1.66.0 by @Kludex in #1202
• Use pydantic-ai-slim version on CLI instead of pydantic-ai by @samuelcolvin in #1213
• Drop system parameter from OpenAIModel by @Kludex in #1235
• Use provider on models inference by @Kludex in #1234
• Add cohere provider class by @hrahmadi71 in #1225

New Contributors

• @derekwsgray made their first contribution in #1227

Full Changelog: v0.0.43...v0.0.44