How to Get a Llama API Key for Free (Where It Is Free, 2026)

How to Get a Llama API Key for Free

Updated 27 October 2026.

In short: Meta's Llama models are open-weight, so other services run them. Llama 3.3 70B is free on Cloudflare Workers AI's daily allowance and on OVHcloud with no API key, and NVIDIA's API catalog has Llama models, including NVIDIA's Llama-based Nemotron models, free after phone verification.

Llama: open-weight models

Meta publishes Llama's weights, so services such as Cloudflare and NVIDIA run the models on their own servers and offer them through their own API keys. "Llama Cloud" in searches is LlamaIndex's product, a different company.

Where Llama models are free

Where	Llama model	Free allowance	What it needs
Cloudflare Workers AI	`@cf/meta/llama-3.3-70b-instruct-fp8-fast`	Inside the 10,000 free Neurons a day	A free Cloudflare account
OVHcloud AI Endpoints	`Meta-Llama-3_3-70B-Instruct`	2 requests a minute per model, per IP address	Nothing: no account and no API key
NVIDIA (build.nvidia.com)	`nvidia/llama-3.1-nemotron-ultra-253b-v1`, `nvidia/llama-3.1-nemotron-70b-instruct`, `meta/codellama-70b`	Up to 40 requests a minute, no daily limit stated	Phone verification, not offered in every country

Llama 3.3 70B is the model in our Cloudflare tutorial's test request. The full list of free AI APIs, with each limit and source, is on how to get AI API keys for free.

Use a free Llama model in an app made for OpenAI

In the app, enter the service's key, its base URL and its Llama model name:

Service	Base URL	Model name
Cloudflare Workers AI	`https://api.cloudflare.com/client/v4/accounts/YOUR_ACCOUNT_ID/ai/v1`	`@cf/meta/llama-3.3-70b-instruct-fp8-fast`
OVHcloud AI Endpoints	`https://oai.endpoints.kepler.ai.cloud.ovh.net/v1`	`Meta-Llama-3_3-70B-Instruct`
NVIDIA	`https://integrate.api.nvidia.com/v1`	`nvidia/llama-3.1-nemotron-ultra-253b-v1`

To put several free keys behind one address on your own computer, the open-source OmniRoute gateway switches to the next key when one runs out.

Words to know

API key: the password an app uses to call a model. See what is an API key.
Open-weight model: a model whose weights are published, so anyone can run it.
Llama: Meta's open-weight model family.
Base URL: the address an app sends its requests to.
Token: a piece of a word; the unit models read and write. See tokens and context windows.

Frequently asked questions

Is there a free Llama API?+

Yes. Llama 3.3 70B is free on Cloudflare Workers AI within the 10,000 free Neurons a day, and NVIDIA's build.nvidia.com catalog has Llama models after phone verification.

What is the Cloudflare model name for Llama?+

@cf/meta/llama-3.3-70b-instruct-fp8-fast, with your Cloudflare account ID in the base URL.

Is Llama Cloud the same as Llama?+

No. Llama Cloud is LlamaIndex's product; Llama is Meta's model family.

Why do other companies offer Llama?+

Meta publishes the model weights, so any service can run Llama on its own servers.