OpenInference Cohere Instrumentation

Python auto-instrumentation library for the Cohere Python client.

Chat, embedding, and rerank calls made with the Cohere v2 client (ClientV2 and AsyncClientV2) are traced and exported as OpenInference spans. Chat spans capture messages, invocation parameters, tool calls, and token counts. Embedding spans capture the model name, input text, invocation parameters, vectors, and token counts. Rerank spans capture the query, model, input documents, ranked output documents, and relevance scores.

Coverage

ClientV2.chat, AsyncClientV2.chat, ClientV2.chat_stream, AsyncClientV2.chat_stream, ClientV2.embed, AsyncClientV2.embed, ClientV2.rerank, and AsyncClientV2.rerank are instrumented. Streamed calls finish their span when the returned iterator is exhausted, with the accumulated output message, tool calls, and token counts.

The following are not traced, and calls to them produce no spans:

• The v1 client (cohere.Client)
• Classify endpoints

Installation

pip install openinference-instrumentation-cohere

PyPI package: openinference-instrumentation-cohere

Quickstart

Install packages needed for this demonstration.

pip install openinference-instrumentation-cohere cohere arize-phoenix opentelemetry-sdk opentelemetry-exporter-otlp

Start Phoenix in the background as a collector. By default, it listens on http://localhost:6006.

phoenix serve

Set up CohereInstrumentor to trace your application and send the traces to Phoenix.

from openinference.instrumentation.cohere import CohereInstrumentor
from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import SimpleSpanProcessor

endpoint = "http://127.0.0.1:6006/v1/traces"
tracer_provider = TracerProvider()
tracer_provider.add_span_processor(SimpleSpanProcessor(OTLPSpanExporter(endpoint)))

CohereInstrumentor().instrument(tracer_provider=tracer_provider)

Run a chat request. Set the CO_API_KEY environment variable with your Cohere API key.

import cohere

co = cohere.ClientV2()
response = co.chat(
    model="command-a-03-2025",
    messages=[{"role": "user", "content": "Why is the sky blue?"}],
)
print(response.message.content[0].text)

More Info

• OpenInference
• Cohere Python client