Skip to main content
Novita AI provides model APIs with OpenAI-compatible chat completions support. Point the OpenAI client at Novita AI’s OpenAI-compatible endpoint and use openinference-instrumentation-openai to send traces to Phoenix.
This page uses Novita AI’s OpenAI-compatible API. There is no separate Novita-specific OpenInference instrumentor required.

Install

Setup

Add your Novita AI API key and the model you want to call as environment variables:
Connect your application to Phoenix with the register function:

Run Novita AI

Observe

Now that you have tracing set up, all invocations of the OpenAI client pointed at Novita AI will be streamed to your running Phoenix for observability and evaluation. Each call appears as a ChatCompletion LLM span with the prompt, response, model name, token usage, and latency.

Common Issues

  • Spans missing: register(..., auto_instrument=True) must run before your application sends requests, and openinference-instrumentation-openai must be installed in the same environment
  • 401 from Novita AI: Use your Novita AI API key, not an OpenAI key
  • Model not found: Confirm NOVITA_MODEL matches a model identifier available in your Novita AI account
  • Base URL: Ensure you’re using https://api.novita.ai/openai as the base URL

Resources

Novita AI LLM API

OpenInference OpenAI Instrumentation