> For the complete documentation index, see [llms.txt](https://docs.apismart.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.apismart.ai/troubleshooting/model-and-request-errors.md).

# Model and Request Errors

Troubleshoot ApiSmart model IDs, unsupported models, invalid request parameters, resolution errors, and model-specific request issues.

Model and request errors usually occur when a model ID is incorrect, a model is used with the wrong endpoint, or request parameters do not meet the selected model's requirements.

Use the checks below before retrying the request.

### 🧩 Check the Model ID

ApiSmart model IDs must match exactly.

Periods (`.`), hyphens (`-`), version numbers, and other characters cannot be changed.

For example:

```
glm-5.1
gpt-5.6-luna
doubao-seedance-2-0-fast
doubao-seedream-5-0
```

Do not replace periods with hyphens or change the version format.

An incorrectly written model ID may result in an unexpected `503`, so always verify the model ID first when a request fails after changing models.

You can retrieve the current model list with:

```bash
curl https://gw.apismart.ai/v1/models \
  -H "Authorization: Bearer <your-api-key>"
```

Use the returned `data[].id` exactly as provided.

### 🔀 Check the Model and Endpoint

A valid model must also support the endpoint you are calling.

For example:

| Error                         | Meaning                                             |
| ----------------------------- | --------------------------------------------------- |
| `403 model_not_allowed`       | The model is not available to your API Key          |
| `403 model_not_image_capable` | The model does not support the Image Generation API |
| `403 model_not_video_capable` | The model does not support the Video Generation API |
| `403 AccessDenied`            | The requested model or operation is not available   |

Note that `GET /v1/models` returns the system model list, but this does not guarantee that every returned model is enabled for your API Key.

For model availability and IDs, see [**Models and Model IDs**](/models-and-pricing/models-and-model-ids.md).

### ⚠️ Check the Request Body

A `400` response usually means the request format or parameters need to be corrected.

Common codes include:

| Code                     | Check                                      |
| ------------------------ | ------------------------------------------ |
| `invalid_request_error`  | Required parameters or request values      |
| `invalid_json`           | JSON syntax and formatting                 |
| `InvalidParameter`       | Model-specific parameter values            |
| `unsupported_resolution` | Resolution supported by the selected model |

Before retrying, check:

* Required parameters such as `model` and input content
* JSON syntax
* Parameter names and value types
* Whether the parameter is supported by the selected model
* Image, video, resolution, and input limits

Do not repeatedly retry the same `400` request without changing the invalid input.

### 🖼️ Image Request Issues

For image generation, check the selected model and image input limits.

Current image inputs support up to **14 reference images**, with a maximum file size of **10 MB per image**. Input and generated images combined must not exceed 15.

The `size` parameter must also use a supported preset or valid pixel dimensions.

For complete requirements, see [**Image Generation API**](/api-guides/image-generation-api.md).

### 🎬 Video Request Issues

Video parameters vary by model.

Parameters such as:

`duration`, `resolution`, `size`, `content`, `ratio`, `image`, `images`, and `input_reference`

may apply differently depending on the selected HappyHorse or Seedance model.

If you receive:

```
400 unsupported_resolution
```

check that the requested resolution is supported by the selected video model.

For model-specific inputs, resolutions, and workflows, see [**Video Generation API**](/api-guides/video-generation-api.md).

### ✅ Quick Troubleshooting Checklist

When a model or request fails:

1. Verify the model ID character by character.
2. Confirm that the model supports the endpoint.
3. Check JSON syntax and required parameters.
4. Review model-specific input and resolution limits.
5. Use `GET /v1/models` to verify the current model ID.
6. Check [**HTTP Status Codes and API Errors**](/troubleshooting/http-status-codes-and-api-errors.md) if the issue continues.

If the problem is related to authentication, balance, or connection errors, see the corresponding troubleshooting guide instead.

### ➡️ Next Steps

* [**Models and Model IDs**](/models-and-pricing/models-and-model-ids.md)
* [**Image Generation API**](/api-guides/image-generation-api.md)
* [**Video Generation API**](/api-guides/video-generation-api.md)
* [**HTTP Status Codes and API Errors**](/troubleshooting/http-status-codes-and-api-errors.md)