> For the complete documentation index, see [llms.txt](https://docs.nexos.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.nexos.ai/gateway-api/models-list.md).

# Models list

The **Models** section provides a centralized view of the AI models available through nexos.ai. It helps you discover models, understand their capabilities and costs, and keep track of upcoming model retirements.

### Where to find it

You can access the Models section from your Gateway platform:

1. Open **nexos.ai Gateway**.
2. In the left sidebar, select **Models**.

You'll see a searchable catalog of all available models with pricing, capabilities, and availability information.

### Two ways to view models

The Models page displays models in two views using tabs at the top of the page:

{% stepper %}
{% step %}

### Company models

Shows the models enabled for your organization. Your company admins manage which models are available. If you need a model that is not currently enabled, contact your company admin to request access.
{% endstep %}

{% step %}

### All nexos.ai models

Shows the full catalog of models available through nexos.ai. The same underlying model may appear multiple times when it's available through different providers or regions. Pricing and other properties may differ between variants.
{% endstep %}
{% endstepper %}

<figure><img src="https://4112003433-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FDcOhapgXnoeDvgGQ8Jm4%2Fuploads%2FxKYwWYXIJyDFQooL073j%2Fnexos_tabs_clickable.gif?alt=media&amp;token=48fdd5f6-6f0e-4b93-a5bf-2f245da9d482" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
**Note:** A model appearing in All nexos.ai models does not mean it's enabled for your organization.&#x20;

Check the Company models tab to see what's available to you.
{% endhint %}

### Finding the right model

Use **Search models** to find a model by name. For more specific requirements, filters at the top of the page let you narrow the catalog by:

* **Modalities** – filter by supported input and output types (text, images, PDFs, audio, video)
* **Capabilities** – find models that support specific functionality (tools, structured outputs, streaming, reasoning)
* **Context** – select the context-window range required for your workload
* **Input pricing** and **Output pricing** – set a price range to find models within your budget
* **More filters** – filter by provider, developer, deployment region, or data-retention requirements (e.g., zero-day retention for privacy-sensitive workloads)

<figure><img src="https://4112003433-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FDcOhapgXnoeDvgGQ8Jm4%2Fuploads%2FjKkIALMBQK5Ccgu6j5Mt%2FRecording%202026-10-28%20105959.gif?alt=media&amp;token=bae74357-827e-43ea-8358-a2aaf5d5084e" alt=""><figcaption></figcaption></figure>

### Understanding the model table

Each row represents a particular model offering. The same model may appear multiple times if available through different providers or regions.

The table can show information such as **Model**, **Provider**, **Region**, **Input price**, **Output price**, **Context**, and **Release date**. Pricing shown in the table represents token pricing for that particular model offering, while **Context** indicates how much information the model can process within a request.

Use **Columns** to choose which information is displayed in the table. Supported columns can also be sorted where applicable, making it easier to compare models by properties such as price, context size, or release date.

### Model status

Each model displays a status indicator to show its availability:

* **Green checkmark** – model is currently available and active for use
* **Other indicators** – show deprecated, retiring soon, or coming soon status

Status helps you identify which models you can use immediately and which may need migration planning.

### Model details page

Click on any model to open its detailed page. This provides additional information about the specific model.

<figure><img src="https://4112003433-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FDcOhapgXnoeDvgGQ8Jm4%2Fuploads%2FJABGYOHK8vjguGHa3nJB%2FRecording%202026-10-28%20110315.gif?alt=media&amp;token=07ad6c8e-16ff-4aa9-99c1-9804d6a67bf7" alt=""><figcaption></figcaption></figure>

At the top, you can see the model's **developer**, **deployment region**, supported **input and output modalities**, available **API endpoints**, and **release date**, together with a short description of the model.

The summary cards provide important operational information:

* **Price In / Out** – The cost of input and output tokens, typically shown per 1 million tokens.
* **Price cache write / read** – Cache-related pricing when it is supported by the model.
* **Context** – The model's maximum context window.
* **Data retention** – How request data is retained for that model offering. For example, **Zero day retention** indicates that the offering is configured without data retention.

### Model Performance

Model performance metrics show how fast the model responds, measured over the last 24 hours:

* **Time to first token** — How long before the model begins responding (P50 is median, P95 is the slow tail)
* **Output speed** — How many tokens per second the model generates (P50 and P95)
* **End-to-end** — Total time from request to complete response (P50 and P95)

These metrics help you understand latency and speed tradeoffs. A model shows metrics once it has enough traffic; new or rarely-used models may not display performance data yet.

<figure><img src="https://4112003433-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FDcOhapgXnoeDvgGQ8Jm4%2Fuploads%2FcL2W6B91eRZPi2GpouV7%2Fimage.png?alt=media&amp;token=55ef4047-056f-4ad1-ab76-c3deae9d44a1" alt=""><figcaption></figcaption></figure>

### Model variants

The **Model variants** section shows models that your team has pre-configured for use within the organization.

A variant can provide a company-specific configuration while still being used like other models. Where configured, **Fallback models** indicate alternative models that can be used as part of that setup.

### Manage model access

#### Requesting a model

If a model you need is not available under **Company models**, contact your company admin or owner to request it be enabled.

#### Permission levels

* **Organization owners** can enable/disable models and contact support to enable locked models
* **Organization members** can view enabled models but cannot change availability

Because model access is managed at the organization level, you can only use models that appear in your **Company models** tab.

### FAQ

**Why do I see the same model multiple times in All nexos.ai models?**\
The same model may be available through different providers, regions, or configurations. Each offering may have different pricing or latency characteristics. Check the Provider and Region columns to understand the differences.

**Can I use a model that's in All nexos.ai models but not in Company models?**\
No. You can only use models enabled in your Company models tab. Contact your organization owner to request access to additional models.

**What does data retention mean?**\
Data retention indicates how your request data is stored. For example, "Zero day retention" means the request data is not retained; your data is processed and discarded immediately. Check this setting if you have privacy requirements.

**What's the difference between a model and a model variant?**\
A model is the base AI offering (e.g., GPT-4). A variant is a company-specific configuration of that model set up by your admins, with custom settings or fallback routing.

**How do I know if a model is about to be retired?**\
Check the Release date and look for retirement date information in the model's detail page. If you see an upcoming retirement date, plan your migration early.

**Can I customize which columns appear in the table?**\
Yes. Click **Columns** to show or hide columns like Input price, Context, Region, or Release date.

**What does the green checkmark status mean?**\
A green checkmark indicates the model is currently available and active. You can use it immediately in your applications and conversations.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.nexos.ai/gateway-api/models-list.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
