Portkey LLM analytics installation

Install dependencies

Required

Full working examples

See the complete Node.js and Python examples on GitHub. If you're using the PostHog SDK wrapper instead of OpenTelemetry, see the Node.js wrapper and Python wrapper examples.

About Portkey

Portkey acts as an AI gateway that routes requests to 250+ LLM providers. The model string format (@integration-slug/model) determines which provider to use, where the slug is the name you chose when setting up the integration in Portkey.

Install the OpenTelemetry SDK, the OpenAI instrumentation, and the OpenAI SDK.

pip install openai opentelemetry-sdk opentelemetry-exporter-otlp-proto-http opentelemetry-instrumentation-openai-v2
npm install openai @posthog/ai @opentelemetry/sdk-node @opentelemetry/resources @opentelemetry/instrumentation-openai

Set up OpenTelemetry tracing

Required

Configure OpenTelemetry to auto-instrument OpenAI SDK calls and export traces to PostHog. PostHog converts gen_ai.* spans into $ai_generation events automatically.

from opentelemetry import trace
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import SimpleSpanProcessor
from opentelemetry.sdk.resources import Resource, SERVICE_NAME
from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter
from opentelemetry.instrumentation.openai_v2 import OpenAIInstrumentor

resource = Resource(attributes={
    SERVICE_NAME: "my-app",
    "posthog.distinct_id": "user_123", # optional: identifies the user in PostHog
    "foo": "bar", # custom properties are passed through
})

exporter = OTLPSpanExporter(
    endpoint="https://us.i.posthog.com/i/v0/ai/otel",
    headers={"Authorization": "Bearer <ph_project_token>"},
)

provider = TracerProvider(resource=resource)
provider.add_span_processor(SimpleSpanProcessor(exporter))
trace.set_tracer_provider(provider)

OpenAIInstrumentor().instrument()
import { NodeSDK, tracing } from '@opentelemetry/sdk-node'
import { resourceFromAttributes } from '@opentelemetry/resources'
import { PostHogTraceExporter } from '@posthog/ai/otel'
import { OpenAIInstrumentation } from '@opentelemetry/instrumentation-openai'

const sdk = new NodeSDK({
  resource: resourceFromAttributes({
    'service.name': 'my-app',
    'posthog.distinct_id': 'user_123', // optional: identifies the user in PostHog
    foo: 'bar', // custom properties are passed through
  }),
  spanProcessors: [
    new tracing.SimpleSpanProcessor(
      new PostHogTraceExporter({
        apiKey: '<ph_project_token>',
        host: 'https://us.i.posthog.com',
      })
    ),
  ],
  instrumentations: [new OpenAIInstrumentation()],
})
sdk.start()

Call Portkey

Required

Now, when you call Portkey with the OpenAI SDK, PostHog automatically captures $ai_generation events via the OpenTelemetry instrumentation.

import openai
from portkey_ai import PORTKEY_GATEWAY_URL

client = openai.OpenAI(
    base_url=PORTKEY_GATEWAY_URL,
    api_key="<portkey_api_key>",
)

response = client.chat.completions.create(
    model="@<integration-slug>/gpt-5-mini",
    messages=[
        {"role": "user", "content": "Tell me a fun fact about hedgehogs"}
    ],
)

print(response.choices[0].message.content)
import OpenAI from 'openai'
import { PORTKEY_GATEWAY_URL } from 'portkey-ai'

const client = new OpenAI({
  baseURL: PORTKEY_GATEWAY_URL,
  apiKey: '<portkey_api_key>',
})

const response = await client.chat.completions.create({
  model: '@<integration-slug>/gpt-5-mini',
  messages: [{ role: 'user', content: 'Tell me a fun fact about hedgehogs' }],
})

console.log(response.choices[0].message.content)

Note: If you want to capture LLM events anonymously, omit the posthog.distinct_id resource attribute. See our docs on anonymous vs identified events to learn more.

You can expect captured $ai_generation events to have the following properties:

Property	Description
`$ai_model`	The specific model, like `gpt-5-mini` or `claude-4-sonnet`
`$ai_latency`	The latency of the LLM call in seconds
`$ai_time_to_first_token`	Time to first token in seconds (streaming only)
`$ai_tools`	Tools and functions available to the LLM
`$ai_input`	List of messages sent to the LLM
`$ai_input_tokens`	The number of tokens in the input (often found in response.usage)
`$ai_output_choices`	List of response choices from the LLM
`$ai_output_tokens`	The number of tokens in the output (often found in `response.usage`)
`$ai_total_cost_usd`	The total cost in USD (input + output)
[...]	See full list of properties

Verify traces and generations
Recommended
Confirm LLM events are being sent to PostHog
Let's make sure LLM events are being captured and sent to PostHog. Under LLM analytics, you should see rows of data appear in the Traces and Generations tabs.

Check for LLM events in PostHog

Next steps

Recommended

Now that you're capturing AI conversations, continue with the resources below to learn what else LLM Analytics enables within the PostHog platform.

Resource	Description
Basics	Learn the basics of how LLM calls become events in PostHog.
Generations	Read about the `$ai_generation` event and its properties.
Traces	Explore the trace hierarchy and how to use it to debug LLM calls.
Spans	Review spans and their role in representing individual operations.
Anaylze LLM performance	Learn how to create dashboards to analyze LLM performance.

Portkey LLM analytics installation

Contents

Install dependencies

Set up OpenTelemetry tracing

Call Portkey

Verify traces and generations

Next steps

Community questions

Was this page useful?