Mistral AI (tracing) | Respan Docs

The Mistral AI Python SDK is the official client for Mistral models. respan-instrumentation-mistralai activates the Mistral AI OpenInference wrapper and normalizes emitted spans to the Respan tracing pipeline.

Set up Respan

Sign up - Create an account at platform.respan.ai
Create an API key - Generate one on the API keys page

Use Respan Gateway

See Mistral AI gateway setup to route Mistral model calls through the Respan gateway.

Example projects

Respan example projects

Setup

Install packages

$ pip install respan-ai respan-instrumentation-mistralai mistralai python-dotenv

Set environment variables

$ export RESPAN_API_KEY="YOUR_RESPAN_API_KEY"
$ export MISTRAL_API_KEY="YOUR_MISTRAL_API_KEY"

Optional:

$ export RESPAN_BASE_URL="https://api.respan.ai/api"
$ export MISTRALAI_MODEL="mistral/mistral-small"

Initialize and run

1 import os
2 
3 from dotenv import load_dotenv
4 from mistralai.client import Mistral
5 from respan import Respan, workflow
6 from respan_instrumentation_mistralai import MistralAIInstrumentor
7 
8 load_dotenv()
9 
10 respan = Respan(
11     api_key=os.environ["RESPAN_API_KEY"],
12     base_url=os.getenv("RESPAN_BASE_URL", "https://api.respan.ai/api"),
13     instrumentations=[MistralAIInstrumentor()],
14 )
15 
16 
17 @workflow(name="mistralai_chat_completion")
18 def run_chat() -> str:
19     with Mistral(api_key=os.environ["MISTRAL_API_KEY"]) as client:
20         response = client.chat.complete(
21             model=os.getenv("MISTRALAI_MODEL", "mistral/mistral-small"),
22             messages=[
23                 {
24                     "role": "user",
25                     "content": "Reply with one concise sentence about tracing.",
26                 }
27             ],
28         )
29         return str(response.choices[0].message.content or "")
30 
31 
32 try:
33     print(run_chat())
34 finally:
35     respan.flush()
36     respan.shutdown()

View your trace

Open the Traces page and search for the workflow name mistralai_chat_completion.

Configuration

Parameter	Type	Default	Description
`**instrumentor_kwargs`	`dict`	`{}`	Extra keyword arguments forwarded to the upstream Mistral AI OpenInference instrumentor.

Attributes

Attach customer identifiers, thread IDs, workflow names, and metadata to Mistral AI calls with propagate_attributes.

1 from respan import propagate_attributes
2 
3 with propagate_attributes(
4     customer_identifier="user_123",
5     thread_identifier="conversation_456",
6     trace_group_identifier="mistralai_support_chat.workflow",
7     metadata={"plan": "pro", "workflow_name": "mistralai_support_chat.workflow"},
8 ):
9     response = client.chat.complete(
10         model="mistral/mistral-small",
11         messages=[{"role": "user", "content": "Summarize our support policy."}],
12     )

Attribute	Type	Description
`customer_identifier`	`str`	Identifies the end user in Respan analytics.
`thread_identifier`	`str`	Groups related messages into a conversation.
`trace_group_identifier`	`str`	Groups spans by workflow name.
`metadata`	`dict`	Custom key-value pairs merged with default metadata.

Examples

Multi-turn chat

1 response = client.chat.complete(
2     model="mistral/mistral-small",
3     messages=[
4         {"role": "user", "content": "Explain tracing in one sentence."},
5         {"role": "assistant", "content": "Tracing records how a request moves through a system."},
6         {"role": "user", "content": "Now make that explanation shorter."},
7     ],
8 )
9 print(response.choices[0].message.content)

Async chat

1 import asyncio
2 import os
3 
4 from mistralai.client import Mistral
5 
6 
7 async def main() -> None:
8     async with Mistral(api_key=os.environ["MISTRAL_API_KEY"]) as client:
9         response = await client.chat.complete_async(
10             model="mistral/mistral-small",
11             messages=[{"role": "user", "content": "Write a short haiku about traces."}],
12         )
13         print(response.choices[0].message.content)
14 
15 
16 asyncio.run(main())