Sentry vs New Relic
Sentry is an error tracker that added tracing; New Relic is an ingest-priced platform. Sentry shows developers the exception and the release behind it, from $26 a month, and can be self-hosted. New Relic takes OTLP natively and charges per GB and per user beyond a free allowance. Teams fixing bugs pick Sentry; teams monitoring whole systems pick New Relic.
Tomasz WierzbickiData & Market Research
Elena Marchetti
Niels AbrahamsenVendors can pay for visibility on this site; it never changes what we say about a product. How that works.
Choose Sentry ifwhat you want first is release-linked error reports for developers, you can live without infrastructure metrics, or exception data has to stay on servers you control.
Choose New Relic ifyou want one platform for a fleet of containers, fed through OpenTelemetry, used daily by a small number of engineers, and you would like to start on the free allowance.
Sentry and New Relic side by side
| Sentry | New Relic | |
|---|---|---|
| Our application performance monitoring rank | #7 of 15 | #11 of 15 |
| Established in | United States | United States |
| Pricing | Free plan for one user; paid plans with error, span and replay quotas, published; self-hosted available | Free tier of 100 GB a month; then per GB ingested plus per core and full-platform user, published |
| List price, 30 September 2026 | Team plan $26 per month billed annually, with 50k errors included and unlimited users | $0.40 per GB ingested beyond 100 GB free per month; full platform users on Standard $10 for the first and $99 for each additional, up to five |
| Free tier | Yes | Yes |
| Open source | — | — |
| Self-hostable | Yes | — |
| Built for | Development teams that want release-linked errors before they want traces | Container fleets where host counting would be punishing |
| Not for | Infrastructure monitoring or a single-pane operations view | Large groups of occasional viewers who all need full access |
Both are reviewed in full in our application performance monitoring ranking: Sentry and New Relic.
Where they differ
Sentry 2 · New Relic 1 · open 1- 01
Catching exceptions per release
Sentry does one thing extremely well: it shows which release broke which user, with the stack trace attached, and the whole product is organised around that. New Relic has an Errors Inbox that groups errors from APM, browser, mobile and Lambda, shows stack traces and counts occurrences per deployment version, but it is one screen inside a large platform. Sentry leads.
Edge: Sentry
- 02
Scope beyond errors
Sentry's performance tracing came later and is thinner than a dedicated APM tool, and infrastructure monitoring is not its job. New Relic lists APM, infrastructure, logs, browser, mobile, synthetic, network and database monitoring on one platform with over 800 integrations, all priced on the data you ingest. A team that needs more than errors will outgrow Sentry's scope sooner.
Edge: New Relic
- 03
Quotas versus gigabytes and seats
Sentry's Team plan is $26 a month on annual billing with 50k errors, 5M spans, 50 replays and unlimited users; more volume is pay-as-you-go. New Relic gives 100 GB a month and one full platform user free, then charges $0.40 per GB, $49 per core user and, on Pro, $349 per full platform user. Sentry's edge is quota, New Relic's is seats.
Depends on the buyer
- 04
Running it on your own servers
Sentry publishes a self-hosted edition, free of licence fees and deployed with Docker Compose, for teams that cannot send exception payloads to an outside party; it comes without dedicated support. New Relic's documentation describes hosted data centre regions in the United States, the European Union and Japan. Where the requirement is your own hardware, Sentry is the one that meets it.
Edge: Sentry
When neither fits
Neither is ideal if you want errors plus real user monitoring from one small vendor, or full tracing on your own cluster. Raygun combines crash reporting, real user monitoring and APM, while SigNoz is OpenTelemetry-first tracing that you can install and host yourself.
Sentry or New Relic: questions
4 answersShould I use Sentry or New Relic for application monitoring?
Use Sentry if the priority is knowing about exceptions and the release that caused them. Use New Relic if you need a wider platform for a container fleet with OpenTelemetry input.
Sentry's tracing is lighter than a dedicated APM product and it has no infrastructure metrics.
How do Sentry and New Relic charge?
Sentry sells plans with included quotas: Team at $26 and Business at $80 a month on annual billing, both with unlimited users.
New Relic charges $0.40 per GB beyond 100 GB free each month, plus $49 per core user and $349 per full platform user on Pro; basic users are free.
Does Sentry or New Relic support OpenTelemetry better?
New Relic has a native OTLP endpoint and states that it ingests all OTLP data, covering traces, metrics and logs.
Sentry accepts OTLP traces and logs in an open beta and says it does not support OTLP metrics. If OpenTelemetry is central to your setup, New Relic is further along.
Is there a free plan for Sentry or New Relic?
Both have one. Sentry's Developer plan is free for a single user with 5k errors, 5M spans and 50 replays.
New Relic's free tier includes 100 GB of ingest a month, unlimited basic users and one full platform user. Self-hosted Sentry also carries no licence fee.
Sources
Checked 30 September 2026- Sentry pricing page: plan prices, included quotas and user limits.
- Sentry self-hosted documentation: the free self-hosted edition, Docker Compose deployment and support terms.
- Sentry OTLP documentation: OTLP traces and logs in open beta, no OTLP metrics.
- New Relic pricing page: free allowance, per-GB rate and user prices per edition.
- New Relic Errors Inbox documentation: error grouping, stack traces, sources and counts per deployment version.
- New Relic platform page: the list of platform capabilities and the 800+ integrations figure.
- New Relic OpenTelemetry documentation: native OTLP endpoint and the signals it ingests.
- New Relic data centre documentation: US, EU and Japan regions.