# OORB RunFP vs Foxglove and Rerun

Compare RunFP, Foxglove, and Rerun for recorded robot data, baseline checks, visualization, and release workflows, with sourced pricing and adoption steps.

Canonical: https://oorb.io/comparisons/robot-reliability-vs-visualization/
Maintained by: OORB
Updated: 2026-10-15

## Which workflow do you need?

Use OORB RunFP when you need recorded robot episodes tied to provenance, named baselines, and repeatable release checks. Choose Foxglove or Rerun when their visualization and data-analysis workflows fit your investigation; these tools overlap, and you can use them together.

This is OORB's workflow comparison, based on the vendor sources linked below and our [public RunFP guides](/docs/). Vendor features and pricing were checked on October 15, 2026.

## RunFP vs Foxglove vs Rerun

| Question | OORB RunFP | Foxglove | Rerun |
| --- | --- | --- | --- |
| How do I start with existing MCAP files? | Import recordings as episodes inside an initialized robot workspace | Open MCAP files locally; comparison mode keeps recordings separate on a shared timeline | Open MCAP files in the Viewer or ingest them through the SDK |
| How do I investigate changes? | Compare episodes and inspect available software, model, and behavior findings | Overlay or inspect sources side by side, adjusting their start times | Visualize recorded data and use the SDK's query and transformation workflows |
| What is the release workflow described here? | Set a named baseline, choose a task protocol, and run a CLI evaluation gate | Use the comparison and analysis facilities your team has configured | Build an evaluation around the SDK and your task's criteria |
| What should I check before adopting it? | Required signals, comparable task conditions, installation access, and outcome definitions | Supported messages, the comparison-mode plan, and team workflows | Supported MCAP decoders and the analysis or ingestion code your team needs |

Foxglove's [comparison-mode documentation](https://docs.foxglove.dev/docs/visualization/comparison-mode) describes separate sources, time alignment, and overlaid or side-by-side inspection. Rerun documents [MCAP ingestion](https://rerun.io/docs/howto/logging-and-ingestion/mcap) and its [SDK and Hub offerings](https://rerun.io/pricing). Those capabilities go beyond a static log viewer.

RunFP's workflow is described in [run comparisons](/docs/compare-robot-runs/), [model provenance](/docs/model-provenance/), and [regression checks](/docs/robot-regression-ci/). This table describes documented workflows; it does not rule out additional integrations or custom evaluation code in another tool.

## Pricing and access

| Product | Published terms checked October 15, 2026 | Source |
| --- | --- | --- |
| Foxglove | Free: $0/month. Pro: $20/month base plus additional usage. Comparison mode is documented for Pro, Enterprise, and Academic plans | [Pricing](https://foxglove.dev/pricing) and [comparison plans](https://docs.foxglove.dev/docs/visualization/comparison-mode) |
| Rerun | SDK: free and open source under Apache-2.0/MIT. Commercial Hub: contact the vendor for deployment pricing | [Pricing and licensing](https://rerun.io/pricing) |
| OORB RunFP | Access is arranged through OORB onboarding; these public guides do not publish a subscription price | [Getting started](/docs/getting-started/) |

The Foxglove base price is not a total deployment quote. Check the current plan and usage terms before buying. We have not run a cost or performance benchmark between these products.

## Who should choose which?

- **Choose Foxglove** when interactive inspection and its recording-comparison workflow address your immediate debugging question.
- **Choose Rerun** when you want its SDK, data model, visualization, and programmatic data workflows in your development loop.
- **Choose RunFP** when you want the documented local episode, provenance, baseline, and CLI-gate workflow for your recorded robot tests.
- **Use them together** when your team needs interactive investigation and a repeatable release decision from the same recordings.

An attractive plot or a passing comparison still needs a physical task definition. See [recorded outcomes and confidence intervals](/docs/robot-success-rate/) before treating telemetry similarity as task success.

## Add RunFP to an existing MCAP workflow

Keep the original recordings and any useful viewer layouts. After obtaining an OORB-provided installation, initialize a workspace in the robot project:

```sh
runfp init --name robot-evaluation --root .
runfp snapshot recordings/baseline.mcap --label pick-place --json
runfp snapshot recordings/candidate.mcap --label pick-place --json
```

Use the returned episode IDs with your prepared task protocol:

```sh
runfp baseline set <baseline-episode-id> --name pick-place
runfp diff <baseline-episode-id> <candidate-episode-id> --protocol protocols/pick-place.yaml --json
```

Review findings against your existing analysis before enabling a gate. This imports recorded evidence; it does not transfer Foxglove layouts, Rerun blueprints, cloud accounts, or a custom evaluation implementation.

## Can I keep using my current viewer?

Yes. You can inspect the original MCAP recordings in your current tool while using RunFP's imported episodes for the documented comparison workflow. Confirm that both analyses use the intended task interval and signals.

[Request RunFP access with your existing recordings](/#pilot-application).