CrewAI

San Francisco, CA · 11–50 employees

Unclaimed profile

1 Documents

1 Evidenced claims

5 Asserted claims

1 Topics

Topics

CrewAI benchmark

Role-based multi-agent crews outperform single-agent approaches by 40–60% on complex multi-step tasks requiring diverse tool access.

Download with company name only. Read

tool calling protocol

Custom tool decorator protocol, MCP via adapter.

asserted Source →

eval harness maturity

Benchmark suite for multi-agent task completion.

asserted Source →

model agnosticism

Supports major providers via LiteLLM integration.

asserted Source →

license obligations

Apache-2.0 for core framework.

evidenced Source →

documentation quality

Growing docs with examples and use-case templates.

asserted Source →

integration breadth

LiteLLM, LangChain tools, and custom tool integrations.

asserted Source →