# DeepSeek

> Connect DeepSeek's AI models to generate text and power your automations.

App · AI Models · 1 action. Web page: https://docs.lodol.com/docs/integrations/deepseek

## Actions

### Generate a DeepSeek response

Sends a prompt to a DeepSeek model and returns the generated text.

**Inputs**

- Prompt (`prompt`, text, required): The message or question to send to DeepSeek.
- Model (`model`, text): Choose which DeepSeek model answers your prompt. Flash is fast and low-cost; Pro is more capable for harder tasks. One of: `deepseek-v4-flash` (DeepSeek V4 Flash (fast & low-cost, recommended)), `deepseek-v4-pro` (DeepSeek V4 Pro (most capable)). Default: `deepseek-v4-flash`.

**Outputs**

- `model` (text): The DeepSeek model that produced the response.
- `response_text` (text): The text that DeepSeek generated for your prompt.

**Example output**

```json
{
  "model": "deepseek-v4-flash",
  "response_text": "Here is a thoughtful response to your question about artificial intelligence and its applications in modern software development."
}
```