Skip to main content
This guide provides instructions for integrating Pydantic Logfire with the TrueFoundry AI Gateway to export OpenTelemetry traces and metrics.

​
What is Pydantic Logfire?

Pydantic Logfire is an observability platform built on OpenTelemetry that accepts OTLP traces and metrics over HTTP. It provides an LLM-aware interface that automatically detects GenAI semantic conventions and renders conversation replays, token usage, and cost breakdowns for AI workloads. It is offered as a managed cloud service with US and EU regions.

​
Key Features of Pydantic Logfire

  • LLM & AI Observability: Automatically detects GenAI span attributes and renders conversation replay, token usage, and cost data — no custom configuration needed.
  • OpenTelemetry Native: Built on the OTLP standard, meaning any OTLP-compatible exporter can send traces and metrics to Logfire without using the Logfire SDK.
  • Multi-region Ingestion: Provides dedicated ingestion endpoints for the US and EU regions to meet data residency requirements.

​
Prerequisites

Before integrating Pydantic Logfire with TrueFoundry, ensure you have:
  1. TrueFoundry Account: Create a TrueFoundry account and follow the instructions in our Gateway Quick Start Guide.
  2. Pydantic Logfire Account: Sign up at logfire.pydantic.dev and create a project to obtain a write token.

​
Understanding the Logfire Endpoint

Logfire provides two regional base URLs:
RegionBase URL
UShttps://logfire-us.pydantic.dev
EUhttps://logfire-eu.pydantic.dev
You must append the signal path to the base URL yourself — TrueFoundry does not auto-append these paths:
  • Traces: https://logfire-us.pydantic.dev/v1/traces
  • Metrics: https://logfire-us.pydantic.dev/v1/metrics
Choose the region that matches where your Logfire project was created. You can confirm your region from the URL shown in your Logfire dashboard after logging in.

​
Integration Steps

1

Create a Write Token in Logfire

  1. Log into logfire.pydantic.dev and open your project.
  2. Navigate to Project Settings → Write Tokens.
  3. Click Create Write Token, give it a name, and copy the token value immediately — it is shown only once.
Pydantic Logfire Project Settings page showing the Write Tokens section
Store the write token in a secrets manager. Enter it directly as the Authorization header value in TrueFoundry — the Logfire OTLP endpoint accepts the raw write token without a Bearer prefix.
2

Configure OTEL Export in TrueFoundry

  1. Go to AI Gateway → Controls → Settings in the TrueFoundry dashboard.
  2. Scroll down to the OTEL Config section and click the edit (✏️) button.
TrueFoundry AI Gateway Settings page showing the OTEL Config section
  1. Enable the Otel Traces Exporter Configuration toggle and fill in:
FieldValue
ToggleEnabled
ProtocolHTTP Configuration
Endpointhttps://logfire-us.pydantic.dev/v1/traces
EncodingProto
Header KeyAuthorization
Header Value<your-logfire-write-token>
TrueFoundry OTEL Traces Exporter Configuration showing Logfire endpoint and Authorization header
  1. Enable the Otel Metrics Exporter Configuration toggle and fill in:
FieldValue
ToggleEnabled
ProtocolHTTP Configuration
Endpointhttps://logfire-us.pydantic.dev/v1/metrics
EncodingProto
Header KeyAuthorization
Header Value<your-logfire-write-token>
  1. Click Save to apply the configuration.
If your Logfire project is in the EU region, replace logfire-us.pydantic.dev with logfire-eu.pydantic.dev in both the traces and metrics endpoint fields.
3

Verify the Integration

  1. Make a request through the TrueFoundry AI Gateway.
  2. Log into logfire.pydantic.dev and open your project.
  3. Navigate to the Live view and confirm spans with service.name tfy-llm-gateway are appearing.
Pydantic Logfire Live view showing LLM call traces from TrueFoundry AI Gateway
  1. Open the Dashboards → Usage Overview view to confirm gateway metrics are flowing in.
Pydantic Logfire Usage Overview dashboard showing gateway metrics from TrueFoundry AI Gateway
Logfire automatically recognizes gen_ai.* span attributes emitted by the TrueFoundry gateway and renders them in its LLM panel — showing token usage, model names, and request/response content without any extra configuration.

​
Configuration Reference

ConfigurationValue
Traces Endpoint (US)https://logfire-us.pydantic.dev/v1/traces
Traces Endpoint (EU)https://logfire-eu.pydantic.dev/v1/traces
Metrics Endpoint (US)https://logfire-us.pydantic.dev/v1/metrics
Metrics Endpoint (EU)https://logfire-eu.pydantic.dev/v1/metrics
ProtocolHTTP
EncodingProto
Auth Header KeyAuthorization
Auth Header ValueYour Logfire write token (no Bearer prefix needed)