Comparison

These benchmarks compare agent instantiation patterns across non-Agno frameworks.

PerformanceEval runs the callable separately for each enabled metric: warm-ups first, then runtime measurements, then memory measurements. num_iterations applies to each metric, and the default is 10 additional warm-up calls. Model retries, tools, delegation, and memory extraction can add provider requests beyond the callable count.

Memory measurements use Python’s tracemalloc; they do not measure process RSS, GPU memory, database-server memory, or remote model memory. Record dependency versions, database state, and model settings when comparing results.

ExampleDescription
Autogen InstantiationDemonstrates agent instantiation benchmarking with AutoGen.
Crewai InstantiationDemonstrates agent instantiation benchmarking with CrewAI.
Langgraph InstantiationDemonstrates agent instantiation benchmarking with LangGraph.
Openai Agents InstantiationDemonstrates agent instantiation benchmarking with OpenAI Agents SDK.
Pydantic Ai InstantiationDemonstrates agent instantiation benchmarking with PydanticAI.
Smolagents InstantiationDemonstrates agent instantiation benchmarking with Smolagents.