> ## Documentation Index
> Fetch the complete documentation index at: https://docs.neosantara.xyz/llms.txt
> Use this file to discover all available pages before exploring further.

# Embeddings

> Hasilkan representasi vektor teks untuk pencarian semantik dan Retrieval-Augmented Generation.

Endpoint `/v1/embeddings` mengonversi teks menjadi vektor floating-point berdimensi tinggi untuk klasterisasi, pencarian semantik, dan pipeline RAG.

<CodeGroup>
  ```python Python (OpenAI SDK) icon="python" theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}}
  from openai import OpenAI
  import os

  client = OpenAI(
      base_url="https://api.neosantara.xyz/v1",
      api_key=os.environ["NEOSANTARA_API_KEY"]
  )

  response = client.embeddings.create(
      model="gemini-embedding-001",
      input=[
          "Neosantara adalah AI Gateway Indonesia untuk developer.",
          "Arsitektur latensi rendah mengoptimalkan inferensi."
      ]
  )

  for item in response.data:
      print(f"Indeks: {item.index}, Panjang Vektor: {len(item.embedding)}")
  ```

  ```bash cURL icon="terminal" theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}}
  curl -X POST https://api.neosantara.xyz/v1/embeddings \
    -H "Authorization: Bearer $NEOSANTARA_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "model": "gemini-embedding-001",
      "input": "Pencarian semantik data perbankan."
    }'
  ```
</CodeGroup>

## Parameter Permintaan

| Parameter         | Tipe              | Wajib    | Keterangan                                                                                   |
| :---------------- | :---------------- | :------- | :------------------------------------------------------------------------------------------- |
| `model`           | `string`          | Ya       | Pengidentifikasi model embedding (contoh: `gemini-embedding-001`, `text-embedding-3-small`). |
| `input`           | `string \| array` | Ya       | Teks tunggal atau daftar teks (array) yang ingin di-embed.                                   |
| `dimensions`      | `integer`         | Opsional | Jumlah dimensi vektor keluaran (didukung pada model tertentu).                               |
| `encoding_format` | `string`          | Opsional | Format vektor: `"float"` atau `"base64"`. Default `"float"`.                                 |

## Model Embedding yang Tersedia

| Model ID                 | Provider   | Context Window | Capabilities | Pricing (Input/Output per 1M) |
| :----------------------- | :--------- | :------------- | :----------- | :---------------------------- |
| `gemini-embedding-001`   | Google     | 2k tokens      | Embeddings   | $0.15 / $0                    |
| `nusa-embedding-0001`    | Neosantara | N/A            | Embeddings   | Free / Included               |
| `nv-embed-v1`            | NVIDIA     | 32k tokens     | Embeddings   | Rp 150 / Rp 0                 |
| `text-embedding-3-small` | OpenAI     | 8k tokens      | Embeddings   | $0.02 / $0                    |


## Related topics

- [Chat Completions](/id/gateway/chat-completions.md)
- [LlamaIndex](/id/integrations/llama-index.md)
- [Integrasi LangChain](/id/integrations/langchain.md)