Cohere Command R — Long-Context Tool-Use Model for Agents

Quick Use

1. Sign up at dashboard.cohere.com → copy API key
2. pip install cohere (or npm install cohere-ai)
3. co.chat(model='command-r-plus-08-2024', message='...') — add tools= and documents= for tool use / RAG

Intro

Command R+ is Cohere's flagship enterprise LLM — 128K context, native tool use, multilingual (10+ languages with strong fluency), and specifically tuned for RAG and agent workloads. Pricing slots between GPT-4o-mini and Claude Sonnet, with comparable benchmark performance on tool use and multi-step reasoning. Best for: enterprise agents, multilingual customer-facing AI, RAG pipelines where citations matter. Works with: Cohere API, AWS Bedrock, Azure, Oracle Cloud. Setup time: 2 minutes.

Hello, Command R+

import cohere

co = cohere.Client(os.environ["COHERE_API_KEY"])

response = co.chat(
    model="command-r-plus-08-2024",
    message="Compare LFP and NMC battery chemistries.",
    temperature=0.3,
)

print(response.text)

Native tool use

tools = [{
    "name": "get_weather",
    "description": "Get current weather",
    "parameter_definitions": {
        "city": {"description": "City name", "type": "str", "required": True},
    },
}]

response = co.chat(
    model="command-r-plus-08-2024",
    message="What's the weather in Tokyo and Berlin? Compare.",
    tools=tools,
)

# Loop until no more tool calls
while response.tool_calls:
    tool_results = []
    for tc in response.tool_calls:
        result = call_my_tool(tc.name, tc.parameters)
        tool_results.append({"call": tc, "outputs": [{"result": result}]})

    response = co.chat(
        model="command-r-plus-08-2024",
        message="",
        tools=tools,
        tool_results=tool_results,
    )

print(response.text)

Built-in RAG mode

documents = [
    {"title": "Doc 1", "snippet": "PyTorch is..."},
    {"title": "Doc 2", "snippet": "TensorFlow is..."},
]

response = co.chat(
    model="command-r-plus-08-2024",
    message="Compare PyTorch and TensorFlow",
    documents=documents,
)

print(response.text)
# Output cites docs by ID — see response.citations
for c in response.citations:
    print(f"{c.text} → {c.document_ids}")

Pricing snapshot (vs alternatives)

FAQ

Q: Is Command R free? A: Cohere offers free trial credits on signup. After that, pay-per-token via cohere.com or via AWS Bedrock / Azure with their billing. Free tier suitable for prototyping; production needs a paid plan.

Q: How does Command R+ compare to Claude Sonnet? A: On English benchmarks Sonnet leads slightly. Command R+ is competitive on tool-use and multilingual tasks at lower price. For enterprise / multilingual / RAG-heavy use cases Command R+ is often the better $/quality.

Q: Does Command R support function calling like OpenAI? A: Yes — native tool use is first-class. Schema is similar but uses parameter_definitions instead of parameters. Cohere's SDK normalizes; with raw API account for the format diff.

Source & Thanks

Built by Cohere. Commercial product with free trial.

docs.cohere.com — Command R+ documentation