---
id: newrelic-telemetry-sdk
version: "0.9.0"
license: Apache-2.0
license_treatment: permissive
maintenance: active
---

# newrelic-telemetry-sdk — New Relic Telemetry SDK

License: permissive · Maintenance: active · Downloads: 846.0K/mo

## What it is and what it does

The New Relic Telemetry SDK is a Python library that sends observability data directly to New Relic's ingestion APIs without requiring a full APM agent. It supports four data types: metrics (gauge, count, summary), events (custom business events), logs (application log messages), and spans (distributed traces). You instantiate a client for each data type, construct the appropriate data objects, and call send or send_batch to transmit them to New Relic.

The package depends only on urllib3 for HTTP transport and supports Python 3.9 through 3.14 on both CPython and PyPy. It's useful when you want fine-grained control over what telemetry you collect and send, or when you need to instrument custom code without deploying a full agent. The documentation references New Relic's rate limits for each API, which you should review before high-volume use.

Use it for:

- Send custom business metrics (e.g., user signups, transaction amounts) to New Relic dashboards
- Instrument distributed traces across microservices by creating and sending Span objects
- Log application events and errors directly to New Relic for centralized audit and diagnostics
- Monitor custom application gauges (e.g., queue depth, cache hit rate) without an agent
- Build lightweight observability into serverless or containerized workloads where agent overhead is undesirable

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Sends custom metrics, events, logs, and distributed traces to New Relic for monitoring and observability without requiring a full agent installation.

Yes, if you need lightweight custom telemetry ingestion to New Relic and want to avoid agent overhead. The package is actively maintained, has no known vulnerabilities, and a permissive license. Install it when you're building observability into specific code paths rather than instrumenting an entire application.

## Install

pip install newrelic-telemetry-sdk
uv add newrelic-telemetry-sdk
poetry add newrelic-telemetry-sdk

## Installing newrelic-telemetry-sdk

Before you install: Low friction installation with a single lightweight runtime dependency (urllib3). Actively maintained with recent releases; marked as Alpha status but in use across a substantial user base.

License in practice: Licensed under Apache-2.0 (permissive), allowing use in commercial and proprietary projects with minimal restrictions.

Quickstart:

pip install newrelic-telemetry-sdk

import os
from newrelic_telemetry_sdk import GaugeMetric, MetricClient

metric_client = MetricClient(os.environ["NEW_RELIC_LICENSE_KEY"])
temp = GaugeMetric("temperature", 78.6, {"units": "Fahrenheit"})
response = metric_client.send_batch([temp])
response.raise_for_status()

Requires NEW_RELIC_LICENSE_KEY environment variable set with a valid New Relic account license key.

Verify before relying:

- Whether the package supports async/await patterns for telemetry submission
- Rate-limit handling behavior and retry strategies beyond the documented API limits
- Performance characteristics when sending high-volume telemetry batches

## Package facts

- License: Apache-2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 846.0K/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

new relic telemetry sdk, send metrics to new relic, custom observability python, distributed tracing sdk, application monitoring library, telemetry data ingestion, new relic api client, observability, new-relic, telemetry

[View on SkillFed](https://skillfed.io/packages/newrelic-telemetry-sdk) · [View on PyPI](https://pypi.org/project/newrelic-telemetry-sdk/)