OpenAI-Compatible Gateway

Ion Clock Lab API Docs

A clean OpenAI-compatible API gateway for chat models, image generation, embeddings, text-to-speech, and audio transcription.

Default language: English Base URL: https://llm.ionclocklab.com/v1 Gateway: LiteLLM

Quick start

Every request needs a user API key. Ask the administrator for a limited virtual key. Do not use the LiteLLM master key in applications.

Model List

Open model list

Open model JSON

Base URL

Use this base URL in OpenAI-compatible SDKs and HTTP requests:

https://llm.ionclocklab.com/v1

Authentication

Pass your key in the Authorization header:

Authorization: Bearer YOUR_API_KEY

Chat Completions

Use this endpoint for normal text chat and assistant-style responses.

curl -X POST 'https://llm.ionclocklab.com/v1/chat/completions' \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [
      {"role": "user", "content": "Hello from Ion Clock Lab"}
    ]
  }'

Python example

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://llm.ionclocklab.com/v1",
)

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[
        {"role": "user", "content": "Hello from Ion Clock Lab"}
    ],
)

print(response.choices[0].message.content)

Image Generation

Use this endpoint to generate images. Recommended model: gpt-image-2

curl -sS -X POST 'https://llm.ionclocklab.com/v1/images/generations' \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "model": "gpt-image-2",
    "prompt": "A clean minimal logo for Ion Clock Lab",
    "n": 1,
    "size": "1024x1024"
  }' > /tmp/image_response.json

Embeddings

Use this endpoint to convert text into embedding vectors. Recommended model: text-embedding-3-small

curl -sS -X POST 'https://llm.ionclocklab.com/v1/embeddings' \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "model": "text-embedding-3-small",
    "input": "Ion Clock Lab optical clock research"
  }'

Text to Speech

Use this endpoint to generate audio from text. Recommended model: tts-1

curl -sS -X POST 'https://llm.ionclocklab.com/v1/audio/speech' \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "model": "tts-1",
    "input": "Hello from Ion Clock Lab LiteLLM.",
    "voice": "alloy"
  }' --output /tmp/tts-test.mp3

Audio Transcription

Use this endpoint to transcribe audio into text. Recommended models: whisper-1, gpt-4o-mini-transcribe

curl -sS --location 'https://llm.ionclocklab.com/v1/audio/transcriptions' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  --form 'file=@"/tmp/tts-test.mp3"' \
  --form 'model="whisper-1"'

Models

The public model list page reads a static JSON file generated from the LiteLLM model registry. It exposes model names and types, not API keys.

Recommended starting models

Use case	Model
Chat	gpt-4o-mini
Chat	gpt-4.1-mini
Chat	deepseek-chat
Image	gpt-image-2
Embeddings	text-embedding-3-small
Text to speech	tts-1
Transcription	whisper-1

Common Errors

Error	Likely cause	Fix
401 Unauthorized	Missing or wrong API key	Use a valid LiteLLM virtual key.
Model not found	The key does not allow that model, or the model is not registered.	Check the model list and key permissions.
Upstream provider error	The upstream provider rejected the request.	Check provider key, model name, and request format.