> ## Documentation Index
> Fetch the complete documentation index at: https://docs.adsumnetworks.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Free tier

> Put Adsum IoT Coder to work in your first minute. No key, no account, no card.

Most tools make you choose a provider, paste an API key, and add a card before you can find out whether they help. We cut all of that.

Install Adsum IoT Coder and it just works. Inference runs on a managed model hosted by Adsum Networks, so you can point the agent at your own firmware in the first minute, not the first hour. It is a real working tier, generous enough to scaffold a project and run a full debug loop, not a locked demo.

<Note>
  No key, no account, no card. The free tier is on by default.
</Note>

## Free tier vs your own key

|                 | Free tier                          | Bring your own key                             |
| :-------------- | :--------------------------------- | :--------------------------------------------- |
| **API key**     | Not required                       | Required                                       |
| **Cost to you** | Nothing, the inference is on us    | Your provider's rates                          |
| **Model**       | Managed by Adsum                   | GLM, Claude, DeepSeek, or any compatible model |
| **Best for**    | First run, evaluation, quick fixes | Daily driver, long sessions, model choice      |

## Switching to your own key

When you want a specific model or heavier usage, drop in a key for the GLM Coding Plan, Claude, DeepSeek, or any OpenAI- or Anthropic-compatible endpoint, and the switch is instant: the task you are in keeps running, no restart.

The free tier is token-metered. A "tokens left" indicator decrements with real usage and the balance is tracked and enforced by Adsum. When you reach the limit, a one-click prompt moves you onto your own key and the same task picks up exactly where it left off.

<Card title="Configure a provider" icon="robot" href="/models">
  Recommended models and the exact settings to match the benchmark.
</Card>