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

# Models

`GET /v1/models`

Lists every model the network can serve right now. The list is live, not a catalog: a model appears only while at least one worker is hosting it.

***

## List models

```bash
curl https://api.robincompute.org/v1/models \
  -H "Authorization: Bearer rcompute_live_your_key_here"
```

### Response

```json
{
  "object": "list",
  "data": [
    {
      "id": "llama-3.3-70b",
      "object": "model",
      "created": 1750000000,
      "owned_by": "meta",
      "robincompute": {
        "tier": "max",
        "credits_per_request": 40,
        "credits_per_1k_tokens": 20,
        "active_workers": 12,
        "median_latency_ms": 1840,
        "parameters": "70B",
        "context_window": 128000
      }
    },
    {
      "id": "qwen3-8b",
      "object": "model",
      "created": 1750000000,
      "owned_by": "qwen",
      "robincompute": {
        "tier": "standard",
        "credits_per_request": 8,
        "credits_per_1k_tokens": 4,
        "active_workers": 47,
        "median_latency_ms": 420,
        "parameters": "8B",
        "context_window": 32768
      }
    }
  ]
}
```

Each entry adds a `robincompute` object on top of the standard OpenAI model schema, carrying live network data.

| Field                   | Description                                                          |
| ----------------------- | -------------------------------------------------------------------- |
| `tier`                  | The model's pricing tier: `lite`, `standard`, `pro`, or `max`        |
| `credits_per_request`   | Flat cost covering completions up to roughly 500 output tokens       |
| `credits_per_1k_tokens` | Rate per 1,000 output tokens once a completion runs longer than that |
| `active_workers`        | How many workers are serving this model right now                    |
| `median_latency_ms`     | p50 time-to-first-token across the current worker pool               |
| `parameters`            | Model size, in parameters                                            |
| `context_window`        | Largest context the model accepts, in tokens                         |

***

## Available models

The launch lineup is below. It grows as workers bring new models online and governance approves them.

| Model ID        | Parameters | Context | Tier     | Credits/request |
| --------------- | ---------- | ------- | -------- | --------------- |
| `llama-3.3-70b` | 70B        | 128K    | Max      | 40              |
| `deepseek-r1`   | 70B        | 128K    | Max      | 40              |
| `llama-3.2-27b` | 27B        | 128K    | Pro      | 18              |
| `qwen3-14b`     | 14B        | 32K     | Pro      | 18              |
| `qwen3-8b`      | 8B         | 32K     | Standard | 8               |
| `mistral-7b`    | 7B         | 32K     | Standard | 8               |
| `llama-3.2-3b`  | 3B         | 128K    | Lite     | 2               |
| `qwen3-1.7b`    | 1.7B       | 32K     | Lite     | 2               |

Model IDs are stable. Build against them. Even after a newer model supersedes one on this list, its ID keeps working until a governance vote formally retires it, and retirement always comes with advance notice.

***

## Model availability

`active_workers` shows how much of the network stands behind a model right now. A larger pool generally means faster first tokens and fewer rejected jobs.

Request a model whose worker count has dropped to zero and you get a `503` with a suggested `retry_after`.

During beta, RobinCompute runs its own seed provider pool so every launch model stays reliably available. Beta means early, not unfinished.

***

## Retrieve a specific model

```bash
curl https://api.robincompute.org/v1/models/qwen3-8b \
  -H "Authorization: Bearer rcompute_live_your_key_here"
```

Returns one model object, shaped exactly like an entry in the list response. Use it for a quick read on worker count and latency before you commit a job.


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.robincompute.org/api-reference/models.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.