# `HuggingfaceClient.Inference.Task.TextToImage`
[🔗](https://github.com/huggingface/huggingface_client/blob/v0.1.0/lib/huggingface_client/inference/tasks/tasks.ex#L115)

Text-to-image generation.

## Output types

- `:blob` (default) — raw image bytes
- `:data_url` — base64 data URL string
- `:url` — direct CDN URL (if provider supports it)
- `:json` — full JSON response

## Example

    {:ok, image_bytes} = HuggingfaceClient.text_to_image(client, %{
      model: "stabilityai/stable-diffusion-2",
      inputs: "a picture of a cat",
      parameters: %{negative_prompt: "blurry"}
    })

    File.write!("cat.jpg", image_bytes)

# `run`

```elixir
@spec run(HuggingfaceClient.Client.t(), map()) ::
  {:ok, term()} | {:error, Exception.t()}
```

Runs the `text-to-image` task.

Returns `{:ok, result}` or `{:error, exception}`.

---

*Consult [api-reference.md](api-reference.md) for complete listing*