Skip to main content

OpenAI Python SDK Compatibility

:::tip Looking for the chat model? InertialAI-0.1 is natively OpenAI-compatible — client.chat.completions.create(...) works against https://inertialai.com/api/v1 with no workarounds. Start with the InertialAI-0.1 quickstart. This page covers the older pattern of pointing the OpenAI SDK at the embeddings endpoint. :::

For new Python integrations against the embeddings endpoint, use the official inertialai-python client.

This page is for teams that already standardize on the OpenAI client and want to point it at InertialAI's embeddings endpoint.

When to use this path​

Use the OpenAI SDK if you already have framework code, shared abstractions, or existing tooling built on openai.

Use the official SDK instead if you want:

  • native EmbeddingMultiModalInput models
  • no manual json.dumps() serialization
  • InertialAI-specific exceptions
  • first-class sync and async clients designed for this API

Installation​

pip install openai

Basic Setup​

Configure the OpenAI client to point to InertialAI's API route:

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://inertialai.com/api/v1",
    api_key=os.environ["INERTIALAI_API_KEY"]
)

Multi-Modal Input Format​

InertialAI's API accepts:

  • Time-series only: Numerical sensor or time-series data without text
  • Text only: Descriptive text without time-series data
  • Multi-modal: Time-series data combined with text for enriched context

Because the OpenAI SDK's embeddings.create() method expects string input, multi-modal payloads must be serialized with json.dumps() before you send them.

import json

embedding_input = [
    {
        "time_series": [
            [34.61, 35.11, 35.21, 35.36, 35.25],  # Channel 1
            [42.13, 42.45, 42.67, 42.89, 43.01]   # Channel 2
        ],
        "text": "Standing jump trial, accelerometer data attached."
    }
]

# Stringify the input for OpenAI SDK compatibility
response = client.embeddings.create(
    model="inertial-embed-alpha",
    input=json.dumps(embedding_input)
)

embedding_vector = response.data[0].embedding

Complete Example​

import json
import os
from openai import OpenAI

client = OpenAI(
    base_url="https://inertialai.com/api/v1",
    api_key=os.environ["INERTIALAI_API_KEY"]
)

embedding_input = [
    {
        "time_series": [
            [34.61, 35.11, 35.21, 35.36, 35.25]
        ],
        "text": "Standing jump trial, accelerometer data attached."
    }
]

response = client.embeddings.create(
    model="inertial-embed-alpha",
    input=json.dumps(embedding_input)
)

embedding = response.data[0].embedding
print(f"Embedding dimensions: {len(embedding)}")
print(f"First 5 values: {embedding[:5]}")
print(f"Tokens used: {response.usage.total_tokens}")

Limitations compared to the official SDK​

Capabilityinertialai-pythonOpenAI SDK
Multi-modal inputs without manual serializationYesNo
Typed EmbeddingMultiModalInput modelsYesNo
InertialAI-specific exceptionsYesNo
Native sync and async clients tailored to InertialAIYesPartial
Compatible with existing OpenAI abstractionsNoYes

Error handling​

from openai import APIError, AuthenticationError, OpenAIError

try:
    response = client.embeddings.create(
        model="inertial-embed-alpha",
        input=json.dumps(embedding_input)
    )
    embeddings = [item.embedding for item in response.data]

except AuthenticationError:
    print("Invalid API key - check INERTIALAI_API_KEY environment variable")

except APIError as e:
    print(f"API error: {e.status_code} - {e.message}")

except OpenAIError as e:
    print(f"OpenAI SDK error: {e}")

Choosing between the two clients​

  • Choose inertialai-python for new Python work.
  • Choose the OpenAI SDK only when compatibility with existing OpenAI-based code matters more than typed InertialAI-native ergonomics.

Next steps​