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

# Models Endpoint

> List available models in OpenAI-compatible API

## Overview

The Models endpoint provides information about available models on the server. It follows the OpenAI API specification.

## Endpoint

```
GET http://localhost:8000/v1/models
```

## Request

No parameters required.

```bash theme={null}
curl http://localhost:8000/v1/models
```

## Response

### Response Body

```json theme={null}
{
  "object": "list",
  "data": [
    {
      "id": "gpt-3.5-turbo",
      "object": "model",
      "created": 1677610602,
      "owned_by": "owner",
      "root": null,
      "parent": null,
      "permission": null
    }
  ]
}
```

### Response Fields

<ResponseField name="object" type="string">
  Always `"list"` for this endpoint
</ResponseField>

<ResponseField name="data" type="array">
  Array of model objects
</ResponseField>

<ResponseField name="data[].id" type="string">
  Model identifier (currently always `"gpt-3.5-turbo"`)
</ResponseField>

<ResponseField name="data[].object" type="string">
  Object type: `"model"`
</ResponseField>

<ResponseField name="data[].created" type="integer">
  Unix timestamp of when the model information was created
</ResponseField>

<ResponseField name="data[].owned_by" type="string">
  Entity that owns the model (default: `"owner"`)
</ResponseField>

<ResponseField name="data[].root" type="string | null">
  Root model identifier (currently `null`)
</ResponseField>

<ResponseField name="data[].parent" type="string | null">
  Parent model identifier (currently `null`)
</ResponseField>

<ResponseField name="data[].permission" type="array | null">
  Model permissions (currently `null`)
</ResponseField>

## Python Example

### Using requests

```python theme={null}
import requests

response = requests.get("http://localhost:8000/v1/models")
models = response.json()

print(f"Available models: {len(models['data'])}")
for model in models['data']:
    print(f"- {model['id']}")
```

### Using OpenAI SDK

```python theme={null}
import openai

# Configure client
openai.api_base = "http://localhost:8000/v1"
openai.api_key = "none"

# List models
models = openai.Model.list()

for model in models.data:
    print(f"Model ID: {model.id}")
    print(f"Created: {model.created}")
    print(f"Owned by: {model.owned_by}")
```

## JavaScript Example

```javascript theme={null}
fetch('http://localhost:8000/v1/models')
  .then(response => response.json())
  .then(data => {
    console.log('Available models:');
    data.data.forEach(model => {
      console.log(`- ${model.id}`);
    });
  });
```

## cURL Example

```bash theme={null}
curl http://localhost:8000/v1/models \
  -H "Content-Type: application/json"
```

## Notes

* The endpoint currently returns a single model entry with ID `"gpt-3.5-turbo"` regardless of which Qwen model is loaded
* This is for OpenAI API compatibility - the actual model serving requests is the Qwen model specified when starting the server
* The `created` timestamp is generated at request time
* The model list does not reflect the actual checkpoint path used to start the server

## Model Information

To determine which Qwen model is actually running, check the server startup logs:

```bash theme={null}
python openai_api.py --checkpoint-path Qwen/Qwen-7B-Chat
```

The server will load the model specified in `--checkpoint-path`, but the API will report it as `"gpt-3.5-turbo"` for compatibility.