Observability for Cohere with Splunk

Step 1

Install the Traceloop SDK and initialize it. It will automatically log all calls to Cohere, with prompts and completions as separate spans.

import boto3
import json

from traceloop.sdk import Traceloop
from traceloop.sdk.decorators import task, workflow

Traceloop.init(app_name="joke_generation_service")
co = cohere.Client(os.environ.get("COHERE_API_KEY"))

import * as traceloop from "@traceloop/node-server-sdk";
import { CohereClient } from "cohere-ai";

traceloop.initialize({
  appName: "sample_cohere",
  apiKey: process.env.TRACELOOP_API_KEY,
  disableBatch: true,
});

const cohere = new CohereClient({
  token: process.env.COHERE_API_KEY ?? "",
});

const sampleGenerate = async () => {
  return await traceloop.withWorkflow({ name: "sample_generate" }, async () => {
    const response = await cohere.generate({
      model: "command-light",
      prompt: "What happened to Pluto?",
      k: 1,
      temperature: 2,
    });

    return response.generations?.[0].text;
  });
};

traceloop.withAssociationProperties({}, async () => {
  console.log(await sampleGenerate());
});

Step 2

Configure your collector to collect traces and send them to Splunk Observability Cloud; point OpenLLMetry to send traces to your Splunk collector

# In your Splunk collector configuration
receivers:
  otlp:
    protocols:
      grpc:
        endpoint: "0.0.0.0:4317"
      http:
        endpoint: "0.0.0.0:4318"
exporters:
  # Traces
  sapm:
    access_token: "${SPLUNK_ACCESS_TOKEN}"
    endpoint: "https://ingest.${SPLUNK_REALM}.signalfx.com/v2/trace"
    sending_queue:
      num_consumers: 32
 pipelines:
  traces:
    receivers: [jaeger, otlp, sapm, zipkin]
    processors:
    - memory_limiter
    - batch
    #- resource/add_environment
    exporters: [sapm]
        
# In the environment of your application
TRACELOOP_BASE_URL=http://splunk-otel-collector:4318

Docs

Observability for Cohere with Splunk

Step 1

Step 2

Discover use cases

Trace prompts and completions in Cohere

More things you can connect to {platform}

Bedrock

LlamaIndex

OpenAI

Watsonx

together.ai

Anthropic

Chroma

Mistral AI

Gemini

LangChain

Ollama