> ## 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.

# Any-LLM

> Standardized Python interface for multi-provider inference, streaming, and model portability.

[Any-LLM](https://mozilla-ai.github.io/any-llm/?utm_source=neosantara-docs\&utm_medium=referral) provides a single, standardized Python interface across diverse AI model providers. Neosantara is officially integrated under provider identifier `neosantara`, delivering uniform streaming semantics, consistent parameter handling, and cross-model portability.

## Installation

Install Any-LLM with the native Neosantara extra:

```bash theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}}
pip install "any-llm-sdk[neosantara]"
```

## Authentication

Export your Neosantara API key in your environment:

```bash theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}}
export NEOSANTARA_API_KEY="nsk_..."
```

The native provider automatically targets `https://api.neosantara.xyz/v1`.

## Native Usage Example

Initialize a reusable client with `AnyLLM.create("neosantara")`:

<CodeGroup>
  ```python Python (Reusable Client) icon="python" theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}}
  from any_llm import AnyLLM

  llm = AnyLLM.create("neosantara")

  response = llm.completion(
      model="gemini-3.8-flash",
      messages=[
          {"role": "user", "content": "Explain the advantages of Indonesia's AI Gateway in one sentence."}
      ]
  )

  print(response.choices[0].message.content)
  ```

  ```python Python (One-Off Completion) icon="python" theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}}
  from any_llm import completion

  response = completion(
      model="neosantara/gemini-3.8-flash",
      messages=[
          {"role": "user", "content": "What is a distributed system?"}
      ]
  )

  print(response.choices[0].message.content)
  ```
</CodeGroup>

## Streaming Responses

Any-LLM supports token streaming directly with the native provider:

```python theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}}
from any_llm import AnyLLM

llm = AnyLLM.create("neosantara")

stream = llm.completion(
    model="deepseek-v4.1-flash",
    messages=[{"role": "user", "content": "Write a quick guide."}],
    stream=True
)

for chunk in stream:
    content = chunk.choices[0].delta.content or ""
    print(content, end="", flush=True)
print()
```

## Any-LLM Integration Advantages

| Capability                  | Manual Wrapper Approach                    | Native Neosantara Provider               |
| :-------------------------- | :----------------------------------------- | :--------------------------------------- |
| **Client Initialization**   | Custom client adapters per invocation      | Instant `AnyLLM.create("neosantara")`    |
| **Code Portability**        | Locked to vendor-specific SDK semantics    | Universal interface across model catalog |
| **Streaming Normalization** | Inconsistent chunk shapes across upstreams | Uniform, standardized chunk streaming    |
| **Environment Management**  | Scattered endpoint configurations          | Centralized via `NEOSANTARA_API_KEY`     |


## Related topics

- [Integrations Overview](/en/integrations/overview.md)
- [LiteLLM](/en/integrations/litellm.md)
- [LangChain Integration](/en/integrations/langchain.md)
- [Model Catalog](/en/gateway/models.md)