# Available Models

Freddy provides access to a curated set of AI models through a unified API. Each model has different capabilities, speeds, and cost profiles.

## Freddy Models

These are Aitronos-native models optimized for the Freddy platform.

| Model | Description | Best For |
|  --- | --- | --- |
| `gpt-4o` | Balanced performance and cost | General-purpose tasks, production use |
| `gpt-4o` | Higher capability, deeper reasoning | Complex analysis, multi-step tasks |
| `gpt-4o` | Optimized for low latency | Real-time applications, simple queries |


`gpt-4o` is the recommended default for most use cases.

## Third-Party Models

You can also access leading AI models from other providers through the same API:

| Provider | Models |
|  --- | --- |
| OpenAI | `gpt-4o`, `gpt-4o-mini` |
| Anthropic | `claude-opus-4`, `claude-sonnet-4`, `claude-haiku-4` |
| Google | `gemini-2.5-pro`, `gemini-2.5-flash` |


Third-party model availability depends on your organization's configuration. Contact your admin if a model is not available to you.

## Specifying a Model

Pass the `model` field in your request body:


```bash
curl https://api.aitronos.com/v1/model/response \
 -H "X-API-Key: $FREDDY_API_KEY" \
 -H "Content-Type: application/json" \
 -d '{
 "organization_id": "org_your_org_id",
 "model": "gpt-4o",
 "inputs": [{"role": "user", "content": "Hello!"}]
 }'
```


```python
response = requests.post(
 "https://api.aitronos.com/v1/model/response",
 headers={"X-API-Key": os.environ["FREDDY_API_KEY"]},
 json={
 "organization_id": "org_your_org_id",
 "model": "gpt-4o", # Change model here
 "inputs": [{"role": "user", "content": "Analyze this dataset..."}],
 },
)
```

If you omit `model`, the API uses `gpt-4o` as the default.

## Assistants and Model Selection

When using an [Assistant](/docs/documentation/core-concepts/assistants), the model is configured on the assistant itself. You can override it per-request by including `model` in the request body — the per-request value takes precedence.

## Synapses and Neurons

Each model consumes [synapses and neurons](/docs/documentation/core-concepts/synapses-and-neurons) differently:

- **Neurons** — Measure input processing (your messages, context, files)
- **Synapses** — Measure output generation (the model's response)


Pro and third-party models generally consume more synapses/neurons per request than speed-optimized models. Check your organization's usage dashboard for consumption details.

## Checking Available Models

To see which models are enabled for your organization, use the Models API:


```bash
curl https://api.aitronos.com/v1/models \
 -H "X-API-Key: $FREDDY_API_KEY"
```

## Related Resources

- [Quick Start](/docs/documentation/getting-started/quick-start) — Your first API call
- [Pricing](/docs/documentation/getting-started/pricing) — Cost per model
- [Create a Response](/docs/api-reference/responses/create) — Full request parameters
- [Synapses and Neurons](/docs/documentation/core-concepts/synapses-and-neurons) — How usage is measured