Use DeepSeek
Connect Maid to DeepSeek's cloud API and access DeepSeek-V3 and the DeepSeek-R1 reasoning model — both competitive with frontier models at a fraction of the cost.
Overview
DeepSeek is a Chinese AI lab that has released a series of high-quality open-weight models, including DeepSeek-V3 (a general-purpose chat model) and DeepSeek-R1 (a reasoning model that shows its chain-of-thought before answering). Both are available via the DeepSeek cloud API.
DeepSeek pricing is notably lower than comparable models from OpenAI or Anthropic, which makes it an attractive option if you do a high volume of queries. The API endpoint is fixed at https://api.deepseek.com and is not configurable.
Maid displays the reasoning trace from DeepSeek-R1 as a collapsible "Show Reasoning" block above the final response — the same way it handles other thinking models.
Get a DeepSeek API key
1
Create an account at platform.deepseek.com.
2
Navigate to API Keys in your dashboard.
3
Click Create API Key and copy the generated key.
Configure Maid
1
Open Settings in Maid and select DeepSeek from the API dropdown.
2
Paste your API key into the API Key field.
3
Maid fetches available models automatically. Select deepseek-chat for general use, or deepseek-reasoner for the R1 reasoning model.
Available models
| Model ID | Description |
|---|---|
| deepseek-chat | DeepSeek-V3 — general-purpose chat and instruction following |
| deepseek-reasoner | DeepSeek-R1 — chain-of-thought reasoning model |
Configuration reference
| Field | Required | Notes |
|---|---|---|
| API Key | Yes | — |
| Model | Yes | Auto-populated |
| Base URL | No | Fixed — not configurable |
| Custom Headers | No | Supported |
| Parameters | No | Provider defaults |