> ## Documentation Index
> Fetch the complete documentation index at: https://veniceai-docs-revamp.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Integración con CrewAI

> Construye sistemas de IA multi-agente con Venice AI y CrewAI

[CrewAI](https://www.crewai.com/) te permite construir sistemas multi-agente autónomos donde agentes de IA especializados colaboran en tareas complejas. Venice AI funciona como un proveedor de LLM drop-in gracias a la compatibilidad con OpenAI.

## Configuración

```bash theme={"dark"}
pip install crewai crewai-tools
```

## Configuración básica

Configura Venice como el proveedor de LLM de CrewAI usando la interfaz compatible con OpenAI:

```python theme={"dark"}
import os

os.environ["OPENAI_API_KEY"] = "your-venice-api-key"
os.environ["OPENAI_API_BASE"] = "https://api.venice.ai/api/v1"
os.environ["OPENAI_MODEL_NAME"] = "venice-uncensored"
```

O configúralo por agente con el objeto LLM:

```python theme={"dark"}
from crewai import LLM

venice_llm = LLM(
    model="openai/venice-uncensored",
    api_key="your-venice-api-key",
    base_url="https://api.venice.ai/api/v1",
    temperature=0.7,
)

# Para tareas de razonamiento complejo
venice_flagship = LLM(
    model="openai/zai-org-glm-5-1",
    api_key="your-venice-api-key",
    base_url="https://api.venice.ai/api/v1",
    temperature=0.3,
)
```

## Tu primer crew

Crea un crew de investigación simple con dos agentes:

```python theme={"dark"}
from crewai import Agent, Task, Crew

# Agente 1: Researcher
researcher = Agent(
    role="Senior Research Analyst",
    goal="Find comprehensive, accurate information on the given topic",
    backstory="You are an expert researcher with a keen eye for detail. "
              "You excel at finding and synthesizing information from multiple sources.",
    llm=venice_flagship,
    verbose=True,
)

# Agente 2: Writer
writer = Agent(
    role="Content Strategist",
    goal="Create engaging, well-structured content from research findings",
    backstory="You are a skilled writer who transforms complex research "
              "into clear, compelling content that readers love.",
    llm=venice_llm,
    verbose=True,
)

# Tarea 1: Research
research_task = Task(
    description="Research the topic: {topic}. "
                "Find key facts, recent developments, and expert opinions. "
                "Provide a structured summary with sources.",
    expected_output="A detailed research summary with key findings, "
                    "organized by subtopic, with at least 5 key points.",
    agent=researcher,
)

# Tarea 2: Escribir el artículo
write_task = Task(
    description="Using the research provided, write a compelling blog post "
                "about {topic}. Include an introduction, main sections, and conclusion.",
    expected_output="A well-written blog post of 500-800 words with clear sections.",
    agent=writer,
    context=[research_task],  # Usa la salida de research_task
)

# Crea y ejecuta el crew
crew = Crew(
    agents=[researcher, writer],
    tasks=[research_task, write_task],
    verbose=True,
)

result = crew.kickoff(inputs={"topic": "The future of privacy-preserving AI"})
print(result)
```

## Crew de análisis de producto multi-agente

Un ejemplo más complejo con agentes especializados:

```python theme={"dark"}
from crewai import Agent, Task, Crew, Process

# Modelos diferentes para distintas capacidades de agente
fast_llm = LLM(model="openai/qwen3-5-9b", api_key="your-key", base_url="https://api.venice.ai/api/v1")
smart_llm = LLM(model="openai/zai-org-glm-5-1", api_key="your-key", base_url="https://api.venice.ai/api/v1")
uncensored_llm = LLM(model="openai/venice-uncensored-1-2", api_key="your-key", base_url="https://api.venice.ai/api/v1")

# Market Analyst — necesita inteligencia
market_analyst = Agent(
    role="Market Research Analyst",
    goal="Analyze market trends and competitive landscape",
    backstory="You are a veteran market analyst with 15 years of experience "
              "in tech markets. You provide unbiased, data-driven insights.",
    llm=smart_llm,
    verbose=True,
)

# Red Team — se beneficia de pensamiento sin censura
red_team = Agent(
    role="Red Team Critic",
    goal="Find weaknesses, risks, and potential failures in business strategies",
    backstory="You are a brutally honest critic who stress-tests ideas. "
              "You find every possible flaw and risk, no matter how uncomfortable.",
    llm=uncensored_llm,  # Sin censura para crítica honesta
    verbose=True,
)

# Strategist — necesita razonamiento
strategist = Agent(
    role="Business Strategist",
    goal="Synthesize analysis into actionable strategy recommendations",
    backstory="You are a McKinsey-trained strategist who creates clear, "
              "actionable plans from complex analyses.",
    llm=smart_llm,
    verbose=True,
)

# Tareas
market_task = Task(
    description="Analyze the market opportunity for: {product_idea}. "
                "Cover market size, competitors, trends, and target audience.",
    expected_output="Structured market analysis with TAM/SAM/SOM estimates, "
                    "top 5 competitors, and 3 key market trends.",
    agent=market_analyst,
)

critique_task = Task(
    description="Critically evaluate this product idea and market analysis. "
                "Find every weakness, risk, and potential failure mode. Be brutally honest.",
    expected_output="A list of at least 5 critical risks, 3 potential failure modes, "
                    "and honest assessment of whether this idea will succeed.",
    agent=red_team,
    context=[market_task],
)

strategy_task = Task(
    description="Based on the market analysis and red team critique, "
                "create a go-to-market strategy that addresses the identified risks.",
    expected_output="A clear go-to-market strategy with: positioning statement, "
                    "3 key differentiators, launch timeline, and risk mitigations.",
    agent=strategist,
    context=[market_task, critique_task],
)

crew = Crew(
    agents=[market_analyst, red_team, strategist],
    tasks=[market_task, critique_task, strategy_task],
    process=Process.sequential,
    verbose=True,
)

result = crew.kickoff(inputs={
    "product_idea": "A privacy-first AI coding assistant that runs on Venice API"
})
print(result)
```

## Usar tools

Mejora los agentes con búsqueda web y otras herramientas:

<Note>
  `SerperDevTool` requiere una variable de entorno `SERPER_API_KEY` de [serper.dev](https://serper.dev). Como alternativa, puedes usar la búsqueda web integrada de Venice pasando `venice_parameters: {"enable_web_search": "auto"}` vía `model_kwargs`, sin necesidad de API key extra. Consulta el apartado [Integración con búsqueda web](/guides/integrations/langchain#web-search-integration) de la guía de LangChain para ver un ejemplo.
</Note>

```python theme={"dark"}
from crewai_tools import SerperDevTool, WebsiteSearchTool
from crewai import Agent, Task, Crew

# Tool de búsqueda web (requiere variable de entorno SERPER_API_KEY)
search_tool = SerperDevTool()

researcher = Agent(
    role="Web Researcher",
    goal="Find the latest information on any topic",
    backstory="You are an expert web researcher.",
    llm=venice_flagship,
    tools=[search_tool],
    verbose=True,
)

task = Task(
    description="Research the latest developments in {topic} from the past week.",
    expected_output="A summary of 5 recent developments with dates and sources.",
    agent=researcher,
)

crew = Crew(agents=[researcher], tasks=[task], verbose=True)
result = crew.kickoff(inputs={"topic": "decentralized AI"})
```

## Guía de selección de modelos para CrewAI

Elige el modelo de Venice adecuado para cada rol de agente:

| Rol del agente                     | Modelo recomendado                                          | Por qué                                                         |
| ---------------------------------- | ----------------------------------------------------------- | --------------------------------------------------------------- |
| Razonamiento complejo / Estrategia | `zai-org-glm-5-1`                                           | El mejor modelo privado de razonamiento                         |
| Análisis sin censura / Red team    | `venice-uncensored-1-2`                                     | Sin filtrado de contenido                                       |
| Tareas rápidas / alto volumen      | `qwen3-5-9b`                                                | El más barato a $0.10/1M tokens de entrada y $0.15/1M de salida |
| Agentes de generación de código    | `qwen3-coder-480b-a35b-instruct`                            | Optimizado para código                                          |
| Tareas multimodales / de visión    | `qwen3-vl-235b-a22b`                                        | Comprensión visual avanzada                                     |
| Equipos con presupuesto ajustado   | `qwen3-5-9b` (rápido) + `venice-uncensored-1-2` (principal) | Combinación de bajo coste                                       |

## Consejos para optimizar costes

1. **Usa modelos más baratos para agentes más simples**: no todos los agentes necesitan un modelo flagship. Usa `qwen3-4b` para formateo, resúmenes o extracciones simples.

2. **Usa `venice-uncensored` para roles creativos/críticos**: es rápido, barato y no rechazará análisis incómodos.

3. **Reserva los modelos flagship para razonamiento**: usa `zai-org-glm-5-1` solo para agentes que necesiten razonamiento complejo o function calling fiable.

4. **Limita las iteraciones máximas**: establece `max_iter` en los agentes para evitar uso descontrolado de tokens:
   ```python theme={"dark"}
   agent = Agent(role="...", goal="...", backstory="...", llm=venice_llm, max_iter=5)
   ```

## Ventaja de privacidad

Las garantías de privacidad de Venice lo hacen ideal para casos de uso de CrewAI que implican:

* **Estrategia empresarial confidencial**: cero retención de datos significa que tu análisis competitivo se mantiene privado
* **Procesamiento de datos sensibles**: los modelos privados nunca registran ni almacenan tus datos
* **Ejercicios de red team**: los modelos sin censura dan feedback honesto sin filtrado de contenido

<CardGroup cols={2}>
  <Card title="Docs de CrewAI" icon="book" href="https://docs.crewai.com/">
    Documentación oficial de CrewAI
  </Card>

  <Card title="Modelos de Venice" icon="database" href="/models/overview">
    Explora todos los modelos de Venice
  </Card>
</CardGroup>