Skip to main content
https://storage.googleapis.com/arize-phoenix-assets/assets/images/phoenix-docs-images/gc.ico

Google Colab

Follow our Colab guide for an interactive example.

MistralAI is a leading provider for state-of-the-art LLMs. The MistralAI SDK can be instrumented using the openinference-instrumentation-mistralai package. Traces can be viewed in Arize.

​
Launch Arize

To get started, sign up for a free Arize account and get your Space ID and API Key.

​
Install

pip install openinference-instrumentation-mistralai mistralai arize-otel

​
API Key Setup

Set the MISTRAL_API_KEY environment variable to authenticate calls made using the SDK. 
export MISTRAL_API_KEY='your_mistral_api_key'

​
Setup

Connect to Arize using the register function and instrument the MistralAI client. 
# Import open-telemetry dependencies
from arize.otel import register

# Setup OTel via our convenience function
tracer_provider = register(
    space_id = "your-space-id", # in app space settings page
    api_key = "your-api-key", # in app space settings page
    project_name = "your-project-name", # name this to whatever you would like
)

# Import openinference instrumentor to map Mistral traces to a standard format
from openinference.instrumentation.mistralai import MistralAIInstrumentor

# Turn on the instrumentor
MistralAIInstrumentor().instrument(tracer_provider=tracer_provider)

​
Run Mistral

Now you can use the MistralAI client as usual. It will automatically pick up the MISTRAL_API_KEY from your environment. 
import os

from mistralai import Mistral
from mistralai.models import UserMessage

api_key = os.environ["MISTRAL_API_KEY"]
model = "mistral-tiny"

client = Mistral(api_key=api_key)

chat_response = client.chat.complete(
    model=model,
    messages=[UserMessage(content="What is the best French cheese?")],
)
print(chat_response.choices[0].message.content)

​
Observe

Now that you have tracing setup, all invocations of Mistral (completions, chat completions, embeddings) will be streamed to your Arize account for observability and evaluation.

​
Resources