DocsGPT/docs/content/Models/local-inference.mdx

---
title: Connecting DocsGPT to Local Inference Engines
description: Connect DocsGPT to local inference engines for running LLMs directly on your hardware.
---

# Connecting DocsGPT to Local Inference Engines

DocsGPT can be configured to leverage local inference engines, allowing you to run Large Language Models directly on your own infrastructure. This approach offers enhanced privacy and control over your LLM processing.

Currently, DocsGPT primarily supports local inference engines that are compatible with the OpenAI API format. This means you can connect DocsGPT to various local LLM servers that mimic the OpenAI API structure.

## Configuration via `.env` file

Setting up a local inference engine with DocsGPT is configured through environment variables in the `.env` file. For a detailed explanation of all settings, please consult the [DocsGPT Settings Guide](/Deploying/DocsGPT-Settings).

To connect to a local inference engine, you will generally need to configure these settings in your `.env` file:

*   **`LLM_PROVIDER`**:  Crucially set this to `openai`. This tells DocsGPT to use the OpenAI-compatible API format for communication, even though the LLM is local.
*   **`LLM_NAME`**:  Specify the model name as recognized by your local inference engine. This might be a model identifier or left as `None` if the engine doesn't require explicit model naming in the API request.
*   **`OPENAI_BASE_URL`**:  This is essential. Set this to the base URL of your local inference engine's API endpoint. This tells DocsGPT where to find your local LLM server.
*   **`API_KEY`**:  Generally, for local inference engines, you can set `API_KEY=None` as authentication is usually not required in local setups.

## Native llama.cpp Support

DocsGPT includes native support for llama.cpp without requiring an OpenAI-compatible server. To use this:

```
LLM_PROVIDER=llama.cpp
LLM_NAME=your-model-name
```

This provider integrates directly with llama.cpp Python bindings.

## Supported Local Inference Engines (OpenAI API Compatible)

DocsGPT is also readily configurable to work with the following local inference engines, all communicating via the OpenAI API format. Here are example `OPENAI_BASE_URL` values for each, based on default setups:

| Inference Engine              | `LLM_PROVIDER` | `OPENAI_BASE_URL`          |
| :---------------------------- | :------------- | :------------------------- |
| LLaMa.cpp (server mode)       | `openai`       | `http://localhost:8000/v1`   |
| Ollama                        | `openai`       | `http://localhost:11434/v1`  |
| Text Generation Inference (TGI)| `openai`      | `http://localhost:8080/v1`   |
| SGLang                        | `openai`       | `http://localhost:30000/v1`  |
| vLLM                          | `openai`       | `http://localhost:8000/v1`   |
| Aphrodite                     | `openai`       | `http://localhost:2242/v1`   |
| FriendliAI                    | `openai`       | `http://localhost:8997/v1`   |
| LMDeploy                      | `openai`       | `http://localhost:23333/v1` |

**Important Note on `localhost` vs `host.docker.internal`:**

The `OPENAI_BASE_URL` examples above use `http://localhost`.  If you are running DocsGPT within Docker and your local inference engine is running on your host machine (outside of Docker), you will likely need to replace `localhost` with `http://host.docker.internal` to ensure Docker can correctly access your host's services. For example, `http://host.docker.internal:11434/v1` for Ollama.

## How the Model Registry Works

DocsGPT uses a **Model Registry** to automatically detect and register available models based on your environment configuration. Understanding this system helps you configure models correctly.

### Automatic Model Detection

When DocsGPT starts, the Model Registry scans your environment variables and automatically registers models from providers that have valid API keys configured:

| Environment Variable   | Provider Models Registered |
| :--------------------- | :------------------------- |
| `OPENAI_API_KEY`       | OpenAI models (gpt-5.1, gpt-5-mini, etc.) |
| `ANTHROPIC_API_KEY`    | Anthropic models (Claude family) |
| `GOOGLE_API_KEY`       | Google models (Gemini family) |
| `GROQ_API_KEY`         | Groq models (Llama, Mixtral) |
| `HUGGINGFACE_API_KEY`  | HuggingFace models |

You can also use the generic `API_KEY` variable with `LLM_PROVIDER` to configure a single provider.

### Custom OpenAI-Compatible Models

When you set `OPENAI_BASE_URL` along with `LLM_PROVIDER=openai` and `LLM_NAME`, the registry automatically creates a custom model entry pointing to your local inference server. This is how local engines like Ollama, vLLM, and others get registered.

### Default Model Selection

The registry determines the default model in this priority order:

1. If `LLM_NAME` is set and matches a registered model, that model becomes the default
2. Otherwise, the first model from the configured `LLM_PROVIDER` is selected
3. If neither is set, the first available model in the registry is used

### Multiple Providers

You can configure multiple API keys simultaneously (e.g., both `OPENAI_API_KEY` and `ANTHROPIC_API_KEY`). The registry will load models from all configured providers, giving users the ability to switch between them in the UI.

## Adding Support for Other Local Engines

While DocsGPT currently focuses on OpenAI API compatible local engines, you can extend its capabilities to support other local inference solutions.  To do this, navigate to the `docsgpt/llm` directory in the DocsGPT repository.  Examine the existing Python files for examples of LLM integrations. You can create a new module for your desired local engine, and then register it in the `llm_creator.py` file within the same directory. This allows for custom integration with a wide range of local LLM servers beyond those listed above.