Observability for BytePlus with Opik

BytePlus is ByteDance’s AI-native enterprise platform offering ModelArk, a comprehensive Platform-as-a-Service (PaaS) solution for deploying and utilizing powerful large language models. It provides access to SkyLark models, DeepSeek V3.1, Kimi-K2, and other cutting-edge AI models with enterprise-grade security and scalability.

This guide explains how to integrate Opik with BytePlus using the OpenAI SDK. BytePlus provides OpenAI-compatible API endpoints that allow you to use the standard OpenAI client with BytePlus models.

Getting started

First, ensure you have both opik and openai packages installed:

pip install opik openai

You’ll also need a BytePlus API key. Find a guide on creating your BytePlus API keys for model services here.

Tracking BytePlus API calls

from opik.integrations.openai import track_openai
from openai import OpenAI
# Initialize the OpenAI client with BytePlus base URL
client = OpenAI(
    base_url="https://ark.ap-southeast.bytepluses.com/api/v3",
    api_key="YOUR_BYTEPLUS_API_KEY"
)
client = track_openai(client)
response = client.chat.completions.create(
    model="kimi-k2-250711",  # You can use any model available on BytePlus
    messages=[
        {"role": "user", "content": "Hello, world!"}
    ],
    temperature=0.7,
    max_tokens=100
)
print(response.choices[0].message.content)

Advanced Usage

Using with @track decorator

You can combine the tracked client with Opik’s @track decorator for comprehensive tracing:

from opik import track
from opik.integrations.openai import track_openai
from openai import OpenAI
client = OpenAI(
    base_url="https://ark.ap-southeast.bytepluses.com/api/v3",
    api_key="YOUR_BYTEPLUS_API_KEY"
)
client = track_openai(client)
@track
def analyze_data_with_ai(query: str):
    """Analyze data using BytePlus AI models."""
    response = client.chat.completions.create(
        model="kimi-k2-250711",
        messages=[
            {"role": "user", "content": query}
        ]
    )
    return response.choices[0].message.content
# Call the tracked function
result = analyze_data_with_ai("Analyze this business data...")

Troubleshooting

Common Issues

1. Authentication Errors: Ensure your API key is correct and has the necessary permissions
2. Model Not Found: Verify the model name is available on BytePlus
3. Rate Limiting: BytePlus may have rate limits; implement appropriate retry logic
4. Base URL Issues: Ensure the base URL is correct for your BytePlus deployment

Getting Help

• Check the BytePlus API documentation for detailed error codes
• Contact BytePlus support for API-specific problems
• Check Opik documentation for tracing and evaluation features

Next Steps

Once you have BytePlus integrated with Opik, you can:

• Evaluate your LLM applications using Opik’s evaluation framework
• Create datasets to test and improve your models
• Set up feedback collection to gather human evaluations
• Monitor performance across different models and configurations

For more information about using Opik with OpenAI-compatible APIs, see the OpenAI integration guide.