PostHog LLM Usage Analytics

https://docs.litellm.ai/docs/observability/posthog_integration
Admin
PostHog LLM Usage Analytics provides specialized AI features to track model usage, performance, and...
Tokens:2,172
Snippets:23
Trust Score:9
Update:4 months ago
Show doc for...
Tokens:
Raw
### Start LiteLLM Proxy

Source: https://docs.litellm.ai/docs/observability/posthog_integration

Start the LiteLLM proxy with the specified configuration file and enable debug mode.

```bash
litellm --config config.yaml --debug  

```

--------------------------------

### LiteLLM Python SDK Quick Start with PostHog

Source: https://docs.litellm.ai/docs/observability/posthog_integration

Initialize LiteLLM with PostHog callbacks and set necessary environment variables for API keys. Make a completion call with user metadata.

```python
import litellm  
import os  
  
# from PostHog  
os.environ["POSTHOG_API_KEY"] = ""  
# Optional, defaults to https://app.posthog.com  
os.environ["POSTHOG_API_URL"] = "" # optional  
  
# LLM API Keys  
os.environ['OPENAI_API_KEY']=""  
  
# set posthog as a callback, litellm will send the data to posthog  
litellm.success_callback = ["posthog"]  
  
# openai call  
response = litellm.completion(  
    model="gpt-3.5-turbo",  
    messages=[  
        {"role": "user", "content": "Hi - i'm openai"}  
    ],  
    metadata = {  
        "user_id": "user-123", # set posthog user ID  
    }  
)  

```

--------------------------------

### Set PostHog Callbacks in LiteLLM Python SDK

Source: https://docs.litellm.ai/docs/observability/posthog_integration

Instantly log responses and errors across all providers by setting `litellm.success_callback` and `litellm.failure_callback` to `["posthog"]`.

```python
litellm.success_callback = ["posthog"]  
litellm.failure_callback = ["posthog"] # logs errors to posthog  

```

--------------------------------

### Configure Custom PostHog Instance

Source: https://docs.litellm.ai/docs/observability/posthog_integration

If using a self-hosted PostHog instance, set the `POSTHOG_API_URL` environment variable to your instance's URL. This allows LiteLLM to send logs to your custom PostHog deployment.

```python
import os
os.environ["POSTHOG_API_URL"] = "https://your-posthog-instance.com"

```

--------------------------------

### LiteLLM Proxy Config for PostHog

Source: https://docs.litellm.ai/docs/observability/posthog_integration

Configure LiteLLM Proxy to use PostHog for success and failure callbacks by specifying them in the `config.yaml` file.

```yaml
model_list:  
  - model_name: gpt-3.5-turbo  
    litellm_params:  
      model: gpt-3.5-turbo  

litellm_settings:  
  success_callback: ["posthog"]  
  failure_callback: ["posthog"]


```

--------------------------------

### Configure Team-Specific PostHog Credentials

Source: https://docs.litellm.ai/docs/observability/posthog_integration

Use the LiteLLM Proxy API to set team-specific PostHog credentials for logging to different PostHog projects.

```bash
curl -X POST 'http://localhost:4000/team/{team_id}/callback' \  
  -H 'Authorization: Bearer sk-1234' \  
  -H 'Content-Type: application/json' \  
  -d '{  
    "callback_name": "posthog",  
    "callback_type": "success",  
    "callback_vars": {  
      "posthog_api_key": "ph_team_specific_key",  
      "posthog_api_url": "https://custom.posthog.com"  
    }  
  }'  

```

--------------------------------

### Set User ID and Custom Metadata with LiteLLM Proxy (OpenAI SDK)

Source: https://docs.litellm.ai/docs/observability/posthog_integration

When using LiteLLM Proxy with the OpenAI Python SDK, pass user ID and custom metadata within the `extra_body` parameter of the `client.chat.completions.create` call.

```python
import openai  

client = openai.OpenAI(  
    api_key="sk-1234",  # Your LiteLLM Proxy API key  
    base_url="http://0.0.0.0:4000"  # Your LiteLLM Proxy URL  
)  

response = client.chat.completions.create(  
    model="gpt-3.5-turbo",  
    messages=[  
        {"role": "user", "content": "Hello world"}  
    ],  
    extra_body={  
        "metadata": {  
            "user_id": "user-123",  # Add user ID for PostHog tracking  
            "project_name": "my-project",  # Add custom metadata  
            "environment": "production"  
        }  
    }  
)  

```

--------------------------------

### Set Environment Variables for PostHog

Source: https://docs.litellm.ai/docs/observability/posthog_integration

Set the `POSTHOG_API_KEY` environment variable to authenticate with PostHog. The `POSTHOG_API_URL` is optional and defaults to the PostHog cloud instance.

```bash
export POSTHOG_API_KEY="your-posthog-api-key"  
# Optional, defaults to https://app.posthog.com  
export POSTHOG_API_URL="https://app.posthog.com" # optional  

```

--------------------------------

### Set User ID and Custom Metadata in LiteLLM Python SDK

Source: https://docs.litellm.ai/docs/observability/posthog_integration

Associate events with specific users and add custom metadata by passing `user_id` and other fields within the `metadata` dictionary during a completion call.

```python
import litellm  

litellm.success_callback = ["posthog"]  

response = litellm.completion(  
    model="gpt-3.5-turbo",  
    messages=[  
        {"role": "user", "content": "Hello world"}  
    ],  
    metadata={  
        "user_id": "user-123",  # Add user ID for PostHog tracking  
        "custom_field": "custom_value"  # Add custom metadata  
    }  
)  

```

--------------------------------

### Set PostHog API Key

Source: https://docs.litellm.ai/docs/observability/posthog_integration

Set the `POSTHOG_API_KEY` environment variable to your PostHog project API key to enable logging. This is a required step for the integration to function.

```python
import os
os.environ["POSTHOG_API_KEY"] = "your-api-key"

```

--------------------------------

### Test LiteLLM Proxy with cURL

Source: https://docs.litellm.ai/docs/observability/posthog_integration

Send a test request to the LiteLLM proxy's chat completions endpoint using cURL. Include model, messages, and metadata for tracking.

```bash
curl --location 'http://0.0.0.0:4000/chat/completions' \  
    --header 'Content-Type: application/json' \  
    --data '{  
    "model": "gpt-3.5-turbo",  
    "messages": [  
        {  
        "role": "user",  
        "content": "what llm are you"  
        }  
    ],  
    "metadata": {  
        "user_id": "user-123",  
        "custom_field": "custom_value"  
    }
}'  

```

--------------------------------

### Disable Logging for Specific Calls

Source: https://docs.litellm.ai/docs/observability/posthog_integration

Use the `no-log` metadata flag to prevent specific LiteLLM completion calls from being logged to PostHog. Ensure `litellm.success_callback` is set to `["posthog"]` for other calls to be logged.

```python
import litellm

litellm.success_callback = ["posthog"]

response = litellm.completion(
    model="gpt-3.5-turbo",
    messages=[
        {"role": "user", "content": "This won't be logged"}
    ],
    metadata={"no-log": True}
)

```

--------------------------------

### Per-Request PostHog Credentials in LiteLLM SDK

Source: https://docs.litellm.ai/docs/observability/posthog_integration

Override PostHog credentials for a specific request by providing `posthog_api_key` and `posthog_api_url` directly in the `litellm.completion` call.

```python
import litellm  

litellm.success_callback = ["posthog"]  

# Use custom PostHog credentials for this specific request  
response = litellm.completion(  
    model="gpt-3.5-turbo",  
    messages=[  
        {"role": "user", "content": "Hello world"}  
    ],  
    posthog_api_key="ph_custom_project_key",  
    posthog_api_url="https://custom.posthog.com"  
)  

```
Tokens:
Raw