Cohere

Use Cohere models with PraisonAI Agents

Environment Variables

Setting MODEL_NAME=cohere/command-r-plus is enough — no need to set OPENAI_BASE_URL. The COHERE_API_KEY environment variable is consulted; OPENAI_API_KEY is not used as a fallback.

If you just want the Cohere default, setting COHERE_API_KEY alone is enough — no MODEL_NAME, no llm=. Agent(...) picks cohere/command-r and routes it through litellm. See Provider Auto-Detection.

import os
from praisonaiagents import Agent

os.environ["MODEL_NAME"] = "cohere/command-r-plus"
os.environ["COHERE_API_KEY"] = "your-cohere-key"

agent = Agent(name="Cohere Agent", instructions="You are powered by Cohere")

Models

• Recommended: command-r-plus (best quality)
• Fast/Cheap: command-r (balanced)
• Latest: command-a-03-2025 (newest)
• Small: command-r7b-12-2024 (efficient)

Python

# export COHERE_API_KEY=your-api-key
from praisonaiagents import Agent

agent = Agent(
    instructions="You are a helpful assistant",
    llm="command-r-plus"
)
agent.start("Explain natural language processing")

With Tools

# export COHERE_API_KEY=your-api-key
from praisonaiagents import Agent

def search_documents(query: str) -> str:
    """Search internal documents."""
    return f"Found documents related to: {query}"

agent = Agent(
    instructions="You are a document assistant",
    llm="command-r",
    tools=[search_documents]
)
agent.start("Find documents about machine learning")

Multi-Agent

# export COHERE_API_KEY=your-api-key
from praisonaiagents import Agent, Task, AgentTeam

researcher = Agent(
    instructions="You research topics thoroughly",
    llm="command-r-plus"
)
writer = Agent(
    instructions="You write clear summaries",
    llm="command-r"
)

task1 = Task(description="Research NLP techniques", agent=researcher)
task2 = Task(description="Write a summary", agent=writer)

agents = AgentTeam(agents=[researcher, writer], tasks=[task1, task2])
agents.start()

CLI

export COHERE_API_KEY=your-api-key

# Basic prompt
python -m praisonai "Explain AI" --llm command-r-plus

# Run agents.yaml
python -m praisonai

YAML

framework: praisonai
topic: NLP research
agents:
  researcher:
    role: NLP Researcher
    goal: Research language processing techniques
    instructions: You are an expert in NLP
    llm:
      model: command-r-plus
    tasks:
      research_task:
        description: Research the latest NLP techniques
        expected_output: Comprehensive NLP research report

  writer:
    role: Technical Writer
    goal: Create clear documentation
    instructions: You write clear technical content
    llm:
      model: command-r
    tasks:
      write_task:
        description: Write a summary of the research
        expected_output: Well-written technical summary