Skip to content
CompactifAI API

Frequently Asked Questions

General Questions

Section titled “General Questions”

What is CompactifAI API?

Section titled “What is CompactifAI API?”

CompactifAI API is an LLM inference service providing access to powerful language models through a simple, standardized API that’s compatible with the OpenAI API standard for unbeatable prices.

Which models are available through the API?

Section titled “Which models are available through the API?”

Visit the Models Catalog for detailed information or call the GET /models endpoint (see API Reference).

We offer two categories of models:

  1. Compressed Models (Optimized using our in-house compression technology for efficiency):

    • cai-llama-3-3-70b-slim
    • cai-mistral-small-3-1-slim
    • hypernova-60b
    • blackstar-10b
    • cai-whisper-large-v3-turbo-slim

  2. Original Models:

    • gpt-oss-20b
    • gpt-oss-120b
    • llama-4-scout
    • llama-3-3-70b
    • llama-3-1-8b
    • mistral-small-3-1
    • whisper-large-v3
    • nemotron-3-nano-omni
    • glm-5-1
    • glm-5-2

How does pricing work?

Section titled “How does pricing work?”

Our API offers a pay-as-you-go pricing model. Pricing varies by model. For the most current pricing, please visit our pricing page.

Technical Questions

Section titled “Technical Questions”

What programming languages can I use with the API?

Section titled “What programming languages can I use with the API?”

Our API is language-agnostic and can be used with any programming language that can make HTTP requests. We provide example code in Python, JavaScript, and cURL, but you can use any other language.

What is the difference between the Chat Completions and Completions endpoints?

Section titled “What is the difference between the Chat Completions and Completions endpoints?”
  • Chat Completions is designed for conversational interfaces. It accepts an array of messages with roles (system, user, assistant) and generates a contextually appropriate response.
  • Completions is designed for text completion tasks. It accepts a single text prompt and generates a continuation of that text.

Does the API include security guardrails for content filtering?

Section titled “Does the API include security guardrails for content filtering?”

Currently, our API does not implement built-in security guardrails or content filtering mechanisms. It is the responsibility of developers to implement appropriate content moderation, safety checks, and guardrails within their own applications when integrating with our API. We recommend implementing client-side filtering and monitoring to ensure compliance with your application’s content policies and your country’s regulations.

Does the API support streaming responses?

Section titled “Does the API support streaming responses?”

Yes, our API supports streaming responses for both Chat Completions and Completions endpoints. Set the stream parameter to true in your request to receive partial results as Server-Sent Events (SSE) as they are generated. This is useful for creating real-time chat interfaces where you want to display the response as it’s being generated rather than waiting for the complete response.

Data Privacy & Processing Location

Section titled “Data Privacy & Processing Location”

Where is my data processed?

Section titled “Where is my data processed?”

Your requests stay inside Multiverse Computing’s private network, hosted with our cloud service provider in EU and US data centers. The API service routes each request straight to an inference engine running the selected model on our own GPUs — your data never leaves that private network, and it’s never shared with third parties.

Can I control which region processes my data?

Section titled “Can I control which region processes my data?”

Yes. Use a regional endpoint if you need requests to stay within a specific geography:

  • EU: https://api-eu.compactif.ai/v1
  • US: https://api-us.compactif.ai/v1

The default global endpoint (https://api.compactif.ai/v1) routes each request to whichever region is fastest at that moment, so a request could land in either. If you don’t have a geography requirement, no changes are needed — keep using the global endpoint.

Can a request be traced back to me?

Section titled “Can a request be traced back to me?”

No. Once you’re authenticated, we strip your user identifier (token or email) from the request and swap in a randomly generated request ID. Everything downstream — logs, metrics — is tagged with that ID only, and we don’t keep any mapping back to who made the request.

Do you train on my data?

Section titled “Do you train on my data?”

Never. Prompts, completions, and other request data are never used to train, fine-tune, or otherwise improve our models. See our Data Privacy page for more details.

How long do you retain my data?

Section titled “How long do you retain my data?”

We don’t retain prompts or completions at all — see Zero Data Retention. The only things we keep are anonymized request logs, used for usage visibility and crash reporting, which are deleted after 30 days, plus minimal billing metadata (model used, input/output token counts). ZDR is enforced end-to-end for the API — nothing that could identify you or your content is stored anywhere in the request path.

Do you use cookies or store personal data (PII)?

Section titled “Do you use cookies or store personal data (PII)?”

No. The API doesn’t use cookies or process personal information about your end users.

Account and API Keys

Section titled “Account and API Keys”

How can I sign up, retrieve an API key?

Section titled “How can I sign up, retrieve an API key?”

For step-by-step instructions on account creation, API key access, see our Authentication page.

What should I do if my API key is compromised?

Section titled “What should I do if my API key is compromised?”

If you suspect that your API key has been compromised, contact us immediately through this form so that we can take appropriate action.

Support and Troubleshooting

Section titled “Support and Troubleshooting”

How do I report issues?

Section titled “How do I report issues?”

For technical issues, please fill in this form with details about the problem you’re experiencing, including any error codes or messages.

How can I rotate my access key?

Section titled “How can I rotate my access key?”

Sign in to the CompactifAI Dashboard, open Manage API Keys, and click Rotate key next to your key.

Is there a status page for the API?

Section titled “Is there a status page for the API?”

Yes, you can check the current status of our API at status.compactif.ai.