Chroma (tracing)

Chroma is an open-source vector database for building semantic search and Retrieval-Augmented Generation systems.

Set up Respan

Create an account at platform.respan.ai and grab an API key .

Paste this into your coding agent:


Help me set up Respan tracing with the Chroma integration.
Follow https://www.respan.ai/docs/integrations/chroma.md

Setup

Install packages

Terminal


pip install respan-ai respan-instrumentation-chroma chromadb

Set environment variables

Terminal


export RESPAN_API_KEY="YOUR_RESPAN_API_KEY"

RESPAN_API_KEY is used to export traces to Respan.

Initialize and run

Python


import os

import chromadb
from respan import Respan, workflow
from respan_instrumentation_chroma import ChromaInstrumentor

respan = Respan(
    api_key=os.environ["RESPAN_API_KEY"],
    instrumentations=[ChromaInstrumentor()],
)


@workflow(name="chroma_query_and_filters_workflow")
def run_chroma_query() -> dict:
    client = chromadb.Client()
    collection = client.get_or_create_collection("respan_docs")

    collection.upsert(
        ids=["doc-1"],
        embeddings=[[0.1, 0.2, 0.3]],
        documents=["Respan traces Chroma operations."],
        metadatas=[{"topic": "observability"}],
    )

    return collection.query(
        query_embeddings=[[0.1, 0.2, 0.3]],
        n_results=1,
        include=["documents", "metadatas", "distances"],
    )


print(run_chroma_query())
respan.shutdown()

View your trace

Open the Traces page and search for workflow name chroma_query_and_filters_workflow.

Configuration

Parameter	Type	Default	Description
`api_key`	`str \| None`	`None`	Falls back to `RESPAN_API_KEY` env var.
`base_url`	`str \| None`	`None`	Falls back to `RESPAN_BASE_URL` env var.
`instrumentations`	`list`	`[]`	Plugin instrumentations to activate (e.g. `ChromaInstrumentor()`).
`customer_identifier`	`str \| None`	`None`	Default customer identifier for all spans.
`metadata`	`dict \| None`	`None`	Default metadata attached to all spans.
`environment`	`str \| None`	`None`	Environment tag (e.g. `"production"`).

Attributes

In Respan()

Python


from respan import Respan
from respan_instrumentation_chroma import ChromaInstrumentor

respan = Respan(
    instrumentations=[ChromaInstrumentor()],
    customer_identifier="user_123",
    metadata={"service": "chroma-vector-db", "version": "1.0.0"},
)

With propagate_attributes

Python


import chromadb

from respan import Respan, propagate_attributes
from respan_instrumentation_chroma import ChromaInstrumentor

respan = Respan(instrumentations=[ChromaInstrumentor()])


def run_query_with_attributes(user_id: str):
    with propagate_attributes(
        customer_identifier=user_id,
        thread_identifier="chroma_thread",
        metadata={"index": "respan_docs"},
    ):
        collection = chromadb.Client().get_or_create_collection("respan_docs")
        return collection.query(query_embeddings=[[0.1, 0.2, 0.3]], n_results=3)

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