Skip to main content
Aptible is running an early beta of our LLM Gateway product, which will help you ensure compliant usage of LLMs, whether you’re using them internally or within your application. To join the beta, let us know that you’re interested here.

​
Features

  • HIPAA compliance through:
    • BAA coverage through Aptible for all models and capabilities
    • Audit logging of all calls/responses to the LLM
    • No retention of PHI for model training purposes
    • Data encryption for all data in transit and at rest
  • Automatic translation to different LLM protocols so you can use Aptible LLM keys anywhere
  • Automatic capacity management behind the scenes for models available through Bedrock

​
Supported models

This initial beta offers an Aptible hosted version of the LLM gateway that allows you to use models offered by Amazon Bedrock, Anthropic, and OpenAI. Examples:
  • openai/gpt-5.2
  • bedrock/qwen.qwen3-coder-480b-a35b-v1:0 (slow 480b parameters model)
  • bedrock/qwen.qwen3-coder-30b-a3b-v1:0 (fast 30b parameters model)

​
How to create and manage LLM gateway keys

Use the Aptible UI to create and manage Bedrock API keys to use with LLM tools or in applications. (Optional first step) Create a dedicated environment for your LLM gateway keys if you want to limit access to specific users. Navigation The following features are available on the LLM Keys page and in the Environment details page → “LLM Keys” tab. Create LLM keys Click “New LLM Key” in the top right to create a new key. Select the environment for the key (as needed) and provide a handle. Then, click “Create LLM Key”.
Screenshot2026 02 20at6 00 09PM
Make sure to copy the key on the next screen — you won’t be able to see the key again once you close this window.
Screenshot2026 02 20at6 00 32PM
See existing LLM gateway keys You can see existing LLM gateway keys by going to the LLM Keys page or the Environment details page → “LLM Keys” tab. Each row shows the key name, status, environment, and creation information.
Screenshot2026 02 20at6 10 57PM
Revoke LLM gateway keys You can revoke LLM keys from the list of keys, either in the LLM Keys page or in the Environment details → LLM Keys tab. Click the red “Revoke” button and “Confirm” to do so.

​
How to use LLM gateway keys

Select a model: To select a model to use your key with, click on an LLM key and go to the Models tab. You’ll see a list of model strings that you can use when configuring your application to use the LLM key.
Screenshot2026 02 20at5 58 10PM
Use LLM Gateway keys to talk to an LLM: …via cURL: 
curl -X POST \\
     -H "Content-Type: application/json" \\
     -H "Authorization: Bearer $YOUR_LLM_GATEWAY_KEY" \\
     -d '{
           "model": "openai/gpt-5.2",
           "messages": [{"role": "user", "content": "Hello world!"}]
         }' \\
     <https://llm-gateway-api.aptible.com/chat/completions>
…via any OpenAI-compatible client 
import openai

client = openai.OpenAI(
    api_key="YOUR_LLM_GATEWAY_KEY",
    base_url="<https://llm-gateway-api.aptible.com>"
)

model_name = "openai/gpt-5.2"

response = client.chat.completions.create(
    model="openai/gpt-5.2",
    messages = [
        {
            "role": "user",
            "content": "Hello world!"
        }
    ]
)

print(response)
Let us know if you have questions or requests regarding other use cases.