CrewAI - Nito

from crewai import LLM, Agent

llm = LLM(
    model="openai/moonshotai/kimi-k3",                       # note the openai/ prefix (see below)
    api_key="z_sk_your_key_here",                  # your Nito API key
    base_url="https://app.trynito.ai/v1",  # Nito's OpenAI-compatible base URL
)

researcher = Agent(
    role="Researcher",
    goal="Answer questions concisely.",
    backstory="A focused analyst.",
    llm=llm,
)

CrewAI model strings are provider/model. The openai/ prefix selects the OpenAI-compatible provider, and base_url and api_key are fields on LLM; what follows the prefix is the catalog ID sent to Nito.

# openai/ = CrewAI provider prefix; z-ai/glm-5.2:confidential = the Nito catalog ID
model="openai/z-ai/glm-5.2:confidential"

Privacy

Use a catalog ID, not an OpenAI name. For attested hardware use a :confidential variant; GET /v1/models lists their :confidential IDs. Encrypted needs client-side encryption and extra headers, which these frameworks do not add. See Privacy Levels.

Framework Integrations

The shared pattern.

OpenAI Compatibility

What differs.

Privacy Levels

The four levels.