### GET /tile/{article_id}/{tile_index}/{chunk_index}

Source: https://pixelrag.ai/docs

Fetch the screenshot for a specific tile using its article ID, tile index, and chunk index.

```APIDOC
## GET /tile/{article_id}/{tile_index}/{chunk_index}

### Description
Fetch the screenshot for a specific tile. Pass the `article_id`, `tile_index`, and `chunk_index` obtained from a search result to get the tile's PNG image.

### Method
GET

### Endpoint
/tile/{article_id}/{tile_index}/{chunk_index}

### Parameters
#### Path Parameters
- **article_id** (integer) - Required - The ID of the article.
- **tile_index** (integer) - Required - The index of the tile.
- **chunk_index** (integer) - Required - The index of the chunk.

### Response
#### Success Response (200)
- **Image** (binary) - The PNG image data for the tile.

### Response Example
(Returns a binary PNG image)
```

--------------------------------

### GET /status

Source: https://pixelrag.ai/docs

Get the current status of the PixelRAG index, including configuration details.

```APIDOC
## GET /status

### Description
Get the index status and configuration of the PixelRAG system.

### Method
GET

### Endpoint
/status

### Parameters
None

### Response
#### Success Response (200)
(Details of the response body are not specified in the source, but it is expected to contain status and configuration information.)
```

--------------------------------

### Fetch Specific Tile Screenshot

Source: https://pixelrag.ai/docs

Retrieve a tile's screenshot using its unique identifiers. This is the final step to get the visual content after a successful search.

```bash
# Fetch a tile screenshot from the results
curl https://pixelrag.ai/api/tile/698618/0/0 --output tile.png
```

--------------------------------

### GET /health

Source: https://pixelrag.ai/docs

Perform a health check on the PixelRAG API.

```APIDOC
## GET /health

### Description
Health check endpoint for the PixelRAG API.

### Method
GET

### Endpoint
/health

### Parameters
None

### Response
#### Success Response (200)
(Typically returns a status indicating the API is healthy, e.g., 'OK' or a JSON object with a health status.)
```

--------------------------------

### Text-Only Search for Tiles

Source: https://pixelrag.ai/docs

Initiate a search using only a text query to find relevant tiles. This is a fundamental step for agents to understand user intent.

```bash
# Text-only search
curl -X POST https://pixelrag.ai/api/search \
  -H "Content-Type: application/json" \
  -d '{"queries": [{"text": "When was the Eiffel Tower built?"}], "n_docs": 5}'
```

--------------------------------

### Search for Tiles with Text Query

Source: https://pixelrag.ai/docs

Use this snippet to find tiles matching a natural-language question. Specify the number of documents to return.

```bash
# 1 — search with a text query
curl -X POST https://pixelrag.ai/api/search \
  -H "Content-Type: application/json" \
  -d '{"queries": [{"text": "How does photosynthesis work?"}], "n_docs": 5}'

# → hits[0] = { "article_id": 5878188, "tile_index": 1, "chunk_index": 0, ... }
```

--------------------------------

### POST /search

Source: https://pixelrag.ai/docs

Search for tiles matching a text query, an image, or a combination of both. Returns a list of matching tiles with their article, tile, and chunk indices.

```APIDOC
## POST /search

### Description
Find matching tiles by sending a natural-language question or an image. Each hit comes back with `article_id`, `tile_index`, and `chunk_index`.

### Method
POST

### Endpoint
/search

### Parameters
#### Request Body
- **queries** (array) - Required - An array of query objects. Each object can contain:
  - **text** (string) - Optional - A natural-language query.
  - **image** (string) - Optional - A base64-encoded image.
- **n_docs** (integer) - Optional - The number of documents to return.

### Request Example
```json
{
  "queries": [
    {
      "text": "How does photosynthesis work?"
    }
  ],
  "n_docs": 5
}
```

### Response
#### Success Response (200)
- **hits** (array) - A list of matching tile objects, each containing:
  - **article_id** (integer) - The ID of the article.
  - **tile_index** (integer) - The index of the tile within the article.
  - **chunk_index** (integer) - The index of the chunk within the tile.

#### Response Example
```json
{
  "hits": [
    {
      "article_id": 5878188,
      "tile_index": 1,
      "chunk_index": 0
    }
  ]
}
```
```

--------------------------------

### Fetch Tile Screenshot

Source: https://pixelrag.ai/docs

Retrieve the screenshot for a specific tile using the article ID, tile index, and chunk index obtained from a search result. The output is saved to a file named 'tile.png'.

```bash
# 2 — fetch that hit's screenshot
curl https://pixelrag.ai/api/tile/5878188/1/0 --output tile.png
```

--------------------------------

### Search for Tiles with Image Query

Source: https://pixelrag.ai/docs

Find tiles that are visually similar to an input image. The image must be base64-encoded and sent inline within the JSON payload.

```bash
# 3 — or search with an image: base64-encode any local file inline
curl -X POST https://pixelrag.ai/api/search \
  -H "Content-Type: application/json" \
  -d "{\"queries\": [{\"image\": \"$(base64 < photo.jpg | tr -d '\n উপেক্ষা')\"}], \"n_docs\": 5}"
```

--------------------------------

### Image-Only Search for Visual Similarity

Source: https://pixelrag.ai/docs

Conduct a search based solely on visual similarity to an input image. The image must be base64 encoded.

```bash
# Image-only search (visual similarity)
curl -X POST https://pixelrag.ai/api/search \
  -H "Content-Type: application/json" \
  -d "{\"queries\": [{\"image\": \"$(base64 < photo.jpg | tr -d '\n উপেক্ষা')\"}], \"n_docs\": 5}"
```

--------------------------------

### Hybrid Search with Text and Image

Source: https://pixelrag.ai/docs

Perform a joint multimodal retrieval by combining both text and image queries in a single request. This is useful for queries like 'what is this X?'.

```bash
# 4 — hybrid: combine text + image in one query (joint multimodal retrieval)
curl -X POST https://pixelrag.ai/api/search \
  -H "Content-Type: application/json" \
  -d "{\"queries\": [{\"text\": \"impressionist oil painting\", \"image\": \"$(base64 < photo.jpg | tr -d '\n উপেক্ষা')\"}], \"n_docs\": 5}"
```

--------------------------------

### Joint Image and Text Search

Source: https://pixelrag.ai/docs

Combine image and text queries for a more precise search, particularly effective for identifying specific objects or scenes described by text.

```bash
# Joint image + text search (best for "what is this X?")
curl -X POST https://pixelrag.ai/api/search \
  -H "Content-Type: application/json" \
  -d "{\"queries\": [{\"image\": \"$(base64 < photo.jpg | tr -d '\n উপেক্ষা')\", \"text\": \"landmark building\"}], \"n_docs\": 5}"
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.