# Google GenAI

Use the `GoogleGenAiSdkInstrumentor` from `opentelemetry-instrumentation-google-genai` to auto-instrument the Google GenAI SDK. Traces are exported to a **local OpenTelemetry Collector** over OTLP/gRPC; the collector forwards to Coralogix.

Before you start

These examples export to a local OpenTelemetry Collector over OTLP/gRPC. Deploy the collector first — see [Code examples → Deploy an OpenTelemetry Collector](https://coralogix.com/docs/user-guides/ai/otel-integration/code-examples/#prerequisites-deploy-an-opentelemetry-collector).

## Install

```bash
pip install google-genai opentelemetry-instrumentation-google-genai opentelemetry-sdk opentelemetry-exporter-otlp-proto-grpc
```

## Environment variables

```bash
export GOOGLE_API_KEY="<YOUR_GOOGLE_API_KEY>"
# Point to your local OTel Collector (which forwards to Coralogix)
export OTEL_EXPORTER_OTLP_ENDPOINT="http://<COLLECTOR_HOST>:4317"
export OTEL_EXPORTER_OTLP_INSECURE="true"
export OTEL_SEMCONV_STABILITY_OPT_IN="gen_ai_latest_experimental"
export OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT="SPAN_AND_EVENT"
```

## Script

```python
import os
from google import genai

# --- OTel imports ---
from opentelemetry import trace
from opentelemetry.exporter.otlp.proto.grpc.trace_exporter import OTLPSpanExporter
from opentelemetry.instrumentation.google_genai import GoogleGenAiSdkInstrumentor
from opentelemetry.sdk.resources import SERVICE_NAME, Resource
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import BatchSpanProcessor


# --- OTel setup: configure tracer provider and OTLP exporter ---
def configure_otel() -> TracerProvider:
    resource = Resource.create({
        SERVICE_NAME: "my-genai-service",
        "cx.application.name": "my-genai-app",
        "cx.subsystem.name": "my-service",
    })
    provider = TracerProvider(resource=resource)
    endpoint = os.getenv("OTEL_EXPORTER_OTLP_ENDPOINT", "http://<COLLECTOR_HOST>:4317")
    insecure = os.getenv("OTEL_EXPORTER_OTLP_INSECURE", "true").lower() == "true"
    provider.add_span_processor(
        BatchSpanProcessor(OTLPSpanExporter(endpoint=endpoint, insecure=insecure))
    )
    trace.set_tracer_provider(provider)
    return provider


def main():
    # OTel: initialize tracing, then auto-instrument the Google GenAI SDK
    provider = configure_otel()
    GoogleGenAiSdkInstrumentor().instrument()

    # Your app logic
    client = genai.Client()
    response = client.models.generate_content(
        model="gemini-2.0-flash",
        contents="What is OpenTelemetry in one sentence?",
    )
    print(f"Gemini response: {response.text}")

    # OTel: flush and shut down the tracer provider
    provider.force_flush()
    provider.shutdown()


if __name__ == "__main__":
    main()
```

## Expected span attributes

- `gen_ai.provider.name` = `"google_genai"`
- `gen_ai.request.model` = `"gemini-2.0-flash"`
- `gen_ai.operation.name` = `"chat"`
- `gen_ai.usage.input_tokens`, `gen_ai.usage.output_tokens`
- `gen_ai.input.messages`, `gen_ai.output.messages` (with content capture)

Tip

Put your Coralogix Send-Your-Data API key on the **collector** (e.g. `CORALOGIX_PRIVATE_KEY` in `otel-collector-config.yaml`), not in the Python exporter. The exporter only needs to reach the local collector.

## Next steps

Look up which open-source library to use for your provider in [Compatibility matrix](https://coralogix.com/docs/user-guides/ai/otel-integration/providers/index.md).