Novita AI
Cloud aggregator for open-source LLMs with OpenAI-compatible APIs, covering DeepSeek, Llama, Qwen, Mistral and more.
Overview
Novita AI provides hosted access to a wide range of open-source language models through a cloud API. You can browse its model library and call selected models with a single Novita API key.
It is useful for teams that want open-source model coverage without self-hosting GPUs, while still keeping an OpenAI-compatible request format.
Novita is an aggregator/hosting platform. Exact quotas, pricing, and available model IDs should always be confirmed in the official console and documentation.
How to Get an API Key
- Register or sign in at novita.ai.
- Open Settings → Key Management and create a new API key.
- Copy the key immediately; it is typically shown only once.
- Call Novita's OpenAI-compatible endpoint with Authorization: Bearer <API Key>.
Common Models
FAQ
What kinds of models does Novita focus on?
It focuses on hosted open-source LLMs such as DeepSeek, Llama, Qwen and Mistral families, with the live catalog defined in the official model library.
Is Novita compatible with the OpenAI SDK?
Yes. Use Novita's documented OpenAI-compatible base URL and your Novita API key.
Does Get Model Key sell Novita credits?
No. This site only links to the official key management page. Billing happens on Novita.