Export data from Helicone

Export Helicone requests, optional bodies and session metadata, then prepare examples for Lunary.

Get started with Team See the steps

$20 per user / month. Payment required to activate your workspace.

Keep the gateway running while you export and validate the request history you need.

Export tooling

Use the documented package or REST API in your account region.

Payloads

Explicitly include bodies when you need inputs and outputs.

Gateway

Treat any routing change as a separate release.

Export options

Export data from Helicone: at a glance
Preserve	Why it matters
Request and session identifiers	Separate individual attempts from user tasks
Request and response bodies	Retain examples usable for debugging or evaluation
Custom properties	Preserve environment, release, and application context
Model, tokens, cost, and timestamps	Compare equivalent behavior and cost units
Feedback	Keep observed ratings separate from approved expected answers

1. Configure the export

Set HELICONE_API_KEY and follow the ETL guide’s current package options. Choose your account region, fixed date and property filters, and a new output file.

Request bodies explicitly when needed.
Preserve session IDs and custom properties.
Record the tool version and filters.

2. Inspect a REST sample

This US endpoint example fetches one page, not a full archive. Use your regional endpoint, add date and application filters, then paginate; custom-property filters belong under request_response_rmt.

Example

curl --fail-with-body --request POST \
  --url https://api.helicone.ai/v1/request/query-clickhouse \
  --header "Content-Type: application/json" \
  --header "Authorization: Bearer $HELICONE_API_KEY" \
  --data '{"filter":"all","limit":100,"offset":0}' \
  --output helicone-first-page.json

3. Validate the result

Reconcile unique request IDs and the time range. Inspect bodies, an error and a full session; preserve cost units and distinguish retries from completed user tasks.

4. Prepare the Lunary dataset

Map selected requests or sessions into the CSV/JSONL dataset format, keeping observed outputs separate from approved expected answers. Retain the raw export and source-ID map, then verify imported counts.

5. Validate instrumentation before cutover

Check one complete application workflow in Lunary. If you also replace the gateway, separately test credentials, model names, streaming, errors and rollback.

Questions & answers

Why are inputs or outputs missing from my export?

Check the selected extraction mode. Helicone's export tooling documents an explicit include-body option; metadata-only extraction is a different result.

Is the REST example a full archive script?

No. It fetches one page to inspect the schema. A complete archive needs explicit filters, pagination, error handling, and count reconciliation.

Does moving observability require replacing Helicone's gateway?

Not as a general rule. Keep gateway dependencies explicit and evaluate any request-path change separately.

By Lunary · Documentation reviewed Sep 16, 2026

Sources & methodology (5)

Based on vendor documentation, not an independent benchmark or hands-on rating. Features and plans can change.

Build with Lunary.

Trace, evaluate, and improve your AI applications.

Get started with Team Request a demo