Observability for Mistral AI with New Relic

OpenLLMetry lets you trace any of the open-source Mistral models. With 5 minutes of work you can get complete view of your system directly into New Relic. See how below.

Step 1

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

from mistralai.client import MistralClient
from mistralai.models.chat_completion import ChatMessage
from traceloop.sdk import Traceloop
from traceloop.sdk.decorators import workflow

Traceloop.init(app_name="joke_generation_service")

client = MistralClient(api_key=os.environ["MISTRAL_API_KEY"])

@workflow(name="joke_creation")
def create_joke():
    response = client.chat(
        model="mistral-tiny",
        messages=[
            ChatMessage(role="user", content="Tell me a joke about OpenTelemetry"),
        ],
    )

    return response.choices[0].message.content

Step 2

Get an access key from New Relic dashboard and set the following environment variables.

TRACELOOP_BASE_URL=https://otlp.nr-data.net:443
TRACELOOP_HEADERS="api-key=<YOUR_NEWRELIC_LICENSE_KEY>"

Docs

Observability for Mistral AI with New Relic

Step 1

Step 2

Discover use cases

More things you can connect to {platform}

LlamaIndex

LangChain

Watsonx

together.ai

Ollama

Bedrock

Anthropic

OpenAI

Cohere

Gemini

Chroma