Projects

Send traces to different projects on Confident AI

Overview

You can specify which project each trace should be sent to by setting the Confident API key on a trace. This is especially useful when you need to separate traces to different projects within the same LLM application.

The Confident API key is unique to each project and can be found in your project settings on Confident AI.

Route Traces to Projects

Python

TypeScript

You can use update_current_trace to set the confident_api_key for a specific trace.

main.py

1 from deepeval.tracing import observe, update_current_trace
2 from openai import OpenAI
3 
4 client = OpenAI()
5 
6 @observe()
7 def llm_app(query: str):
8     res = client.chat.completions.create(
9         model="gpt-4o",
10         messages=[{"role": "user", "content": query}]
11     ).choices[0].message.content
12 
13     if query == "Write me a poem.":
14         update_current_trace(confident_api_key="confident-api-key-1")
15     else:
16         update_current_trace(confident_api_key="confident-api-key-2")
17     return res
18 
19 llm_app("Write me a poem.")

confident_api_key overrides the CONFIDENT_API_KEY environment variable.

1	from deepeval.tracing import observe, update_current_trace
2	from openai import OpenAI
3
4	client = OpenAI()
5
6	@observe()
7	def llm_app(query: str):
8	res = client.chat.completions.create(
9	model="gpt-4o",
10	messages=[{"role": "user", "content": query}]
11	).choices[0].message.content
12
13	if query == "Write me a poem.":
14	update_current_trace(confident_api_key="confident-api-key-1")
15	else:
16	update_current_trace(confident_api_key="confident-api-key-2")
17	return res
18
19	llm_app("Write me a poem.")