Stop guessing which one. Put them side by side.

The grid says what these tools are. This says what builders did about it — one decision per person, changeable whenever you change your mind.

# Tool comparison — Grafana Cloud vs Highlight.io vs Axiom

Source: Kbaise, a directory of tools that work with Lovable projects.
Pick one and tell me why before writing any integration code.

## Grafana Cloud (grafana-cloud)
- URL: https://grafana.com/products/cloud/
- Category: Monitoring & error tracking
- Cost: mixed — Free: 10K metric series, 50GB logs, 50GB traces, 50GB profiles, 14-day retention, 3 users. Pro $19/mo platform fee plus $6.50 per 1K series and $8 per active user.
- What it is: Hosted Prometheus, Loki, Tempo and Pyroscope with Grafana dashboards, synthetic monitoring and alerting on top. The free tier is unusually generous on volume at 50GB each of logs and traces per month. The same components are open source, so you can migrate to a fully self-hosted stack later.
- Why builders pick it: Fifty free gigabytes of logs plus a credible self-hosted escape hatch means you are never locked in on price, and OpenTelemetry ingest works out of the box.
- Trap: The free tier caps you at 3 active users and 14-day retention. Metrics bill by active series, so high-cardinality Prometheus labels inflate the bill the same way they do on Datadog.
- Wiring: docker run -d -p 3000:3000 grafana/grafana
- Full dossier: /api/public/tools/grafana-cloud

## Highlight.io (highlight-io)
- URL: https://www.highlight.io
- Category: Monitoring & error tracking
- Cost: free — Self-hosting is free under Apache 2.0. The hosted cloud ended 28 February 2026 and existing users were migrated to LaunchDarkly Observability.
- What it is: An open-source full-stack monitoring app combining session replay, error monitoring, logging and OpenTelemetry tracing in one self-hostable Docker stack. LaunchDarkly acquired it and folded the features into LaunchDarkly Observability. The GitHub repo remains public and actively committed to, but the hosted app.highlight.io service has shut down.
- Why builders pick it: It is now purely a self-host play: if you want replay plus errors plus traces on your own hardware with no per-event bill, this is one of very few Apache-2.0 options.
- Trap: Do not plan around the cloud product. It ended on 28 February 2026 and SDK snippets had to be repointed at LaunchDarkly before 1 March 2026. Self-hosting means operating ClickHouse, Kafka and Postgres yourself.
- Wiring: git clone https://github.com/highlight/highlight && docker compose up
- Full dossier: /api/public/tools/highlight-io

## Axiom (axiom)
- URL: https://axiom.co
- Category: Monitoring & error tracking
- Cost: mixed — Free Personal: 500GB ingest/mo, 25GB storage, 30-day retention, 10 GB-hours query. Axiom Cloud $25/mo platform fee including 1TB ingest, 100GB storage and 100 GB-hours.
- What it is: A serverless log and event store built on object storage and queried with the APL language, with dashboards, monitors and native OpenTelemetry ingest. The permanently free Personal plan allows 500GB of ingest per month, which is enormous for a solo project. Cloud-only, no self-hosted build.
- Why builders pick it: 500GB a month free is more log volume than a side project will ever generate, and the Vercel and Cloudflare integrations mean you can ship logs without writing a shipper.
- Trap: Free retention is fixed at 30 days and query compute is capped at 10 GB-hours per month, so busy dashboards hit that ceiling long before ingest becomes the constraint. No self-hosting option exists.
- Wiring: npm i @axiomhq/js
- Full dossier: /api/public/tools/axiom

## Quick read

- Free with no card: Highlight.io.

## Rules
1. Prefer the free tier when no budget was stated, and say what the ceiling is.
2. Read the full dossier before integrating.
3. Fetch /api/public/models before writing any AI model ID.

Agents can fetch the same thing: GET /api/public/compare?slugs=grafana-cloud,highlight-io,axiom