Autometa Manual
Is Home
Is Home
Locale
en
Navigation Label
Model Settings
Order
30
Parent
SEO Description
Guide to configuring AI models in Autometa.
SEO Title
Autometa Model Settings
Show in Sidebar
Show in Sidebar
Slug
model-settings
Status
Published
Translation Key
model-settings
Model Settings
To start using ‣autometa, the first thing you need to do is add an AI model. You can add an AI model from the Settings / Pengaturan menu.
To add a model:
- Click “Add model analyzer”
- Choose a model from the provider you want to use
- Enter the API key from the provider you’re using
- Click the wrench icon to configure Rate Limit
- Click “Save Model” to save it
Configuring Rate Limits
This feature is used to set limits on the requests sent to your AI model so they don’t exceed the capacity of the API service you’re using—especially if you use services like OpenAI, Gemini, and others.
Concurrent Requests
The maximum number of requests that can run at the same time. Set how many requests can be processed in parallel.
Rate Limit
You can set the maximum number of requests (per second, per minute, or per hour) that Autometa is allowed to send to your AI model. For example:
- 15 requests per minute means Autometa will send at most 15 requests in one minute; if there’s a 16th request, it will be sent in the next minute.
Rate limit warning
Set rate limits based on your model and subscription tier to avoid API throttling. If you’re using a free API tier, make sure your rate limit does not exceed the allowed quota.
For example, Gemini Free only allows 15 requests per minute.
Multiple API Keys
Autometa supports unlimited multiple API keys. You can connect multiple API keys at once to increase processing capacity.
➡️ Example: If you use Gemini Free with a limit of 15 requests per minute, then with 10 API keys, you can process up to 150 requests per minute in parallel.
Geni Model by ‣autometa
Autometa offers a flexible credit system with the Geni model for those who want to use Gemini Paid without dealing with rate limits—easily and affordably:
- Minimum top up: Rp 20.000 (via QRIS / GoPay / ShopeePay directly in the app)
- Generation cost: around Rp 1.4 per image/video (depends on the number of AI tokens used)
- Your balance will be deducted automatically after use, and you can view usage history for the last 10 days in the app
- If you use the Geni model, you don’t need to configure rate limits—Geni by ‣autometa has no rate limit
Geni model preview
OpenAI Model
To get an OpenAI API key:
- Open https://platform.openai.com/api-keys and sign in to your OpenAI account to create/get an API key
- If you don’t have an API key yet, click “Create new secret key” to create one
- Copy the API key and paste it into the API Key field in ‣autometa
For token pricing, see Pricing.
For rate limits, see OpenAI Platform.
Gemini Model
To get a Gemini API key:
- Open https://aistudio.google.com/app/apikey and sign in with your Google account
- Click “Create API Key” if you don’t have one yet
- Copy the API key and enter it into the API Key field in ‣autometa
For Gemini rate limits, see Rate limits | Gemini API | Google AI for Developers
Claude (Anthropic) Model
To get a Claude API key:
- Visit https://console.anthropic.com/account/keys and sign in to your Anthropic account
- Click “Create Key” to create a new API key
- Copy the API key and paste it into the API Key field in ‣autometa
For Claude rate limits, see Rate limits - Anthropic
Ollama (Local) Model
To use Ollama as an AI model:
- Make sure you have installed Ollama from https://ollama.com (available for macOS, Linux, and Windows WSL)
- Run Ollama in the background (it typically starts automatically after installation)
- No API key is needed because the model runs locally
Custom Model (OpenAI-Compatible)
Autometa also supports other AI model providers that are compatible with the OpenAI API, such as:
Including local AI options such as:
Steps:
- Select the “Custom” model in ‣autometa
- Enter the endpoint URL for your provider, for example:
- Create an account and API key from the provider (e.g., Fireworks or Nebius)
- For local AI, the API key field can be left blank
- Copy the model name into the Model field in ‣autometa
Tutorial video
