Skip to content

Open-weight models through an OpenAI-compatible API.

Serverless, so there’s nothing to provision or scale: every model sits behind one endpoint and one key. Pick the model per call and pay per token at published rates. Prompts are never stored.

Get an API key

Featured models

Billed per token, per model. See rates →

  • DeepSeek

    DeepSeek V4.1 Flash

    deepseek/deepseek-v4.1-flash

    Context
    1M tokens
    Input
    Text, images

  • DeepSeek

    DeepSeek V4 Pro

    deepseek/deepseek-v4-pro

    Context
    1M tokens
    Input
    Text

  • Z.ai

    GLM 5.2

    z-ai/glm-5.2

    Context
    256K tokens
    Input
    Text

  • Moonshot AI

    Kimi K3

    moonshotai/kimi-k3

    Context
    1M tokens
    Input
    Text, images

Already using the OpenAI SDK?

Change the base URL and the key, and the rest of your code stays as it is. To switch models, change the model name.

Create a key ↗Read the docs ↗

chat.py
import osfrom openai import OpenAIclient = OpenAI(    base_url="https://ai.aur.lu/v1",    api_key=os.environ["AURORA_API_TOKEN"],)reply = client.chat.completions.create(    model="deepseek/deepseek-v4.1-flash",    messages=[{"role": "user", "content": "Summarize this contract."}],)print(reply.choices[0].message.content)

Dedicated models

When traffic is steady or the data can’t share hardware, run your model on GPUs reserved for you. Aurora deploys and operates the model you choose, open-weight or your own fine-tune, on single-tenant NVIDIA systems, with capacity and availability written into your agreement.

Need a model for your company alone? Talk to us →

Open the portal. See it for yourself.

Create an account, make a bucket, and send your first inference call in minutes. No sales call, no commitment.

Start freeConnect with us

30-day free trial · no payment method required