https://github.com/sassanix/tgptopnwebuifunction
Allows for tgpt to be interacted within openwebui.
https://github.com/sassanix/tgptopnwebuifunction
Last synced: about 1 month ago
JSON representation
Allows for tgpt to be interacted within openwebui.
- Host: GitHub
- URL: https://github.com/sassanix/tgptopnwebuifunction
- Owner: sassanix
- Created: 2025-02-27T18:38:39.000Z (2 months ago)
- Default Branch: main
- Last Pushed: 2025-02-27T18:45:02.000Z (2 months ago)
- Last Synced: 2025-02-28T02:56:45.445Z (2 months ago)
- Size: 8.79 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# 🔌 tgpt-open-webui-integration
**Connect tgpt to Open WebUI for Enhanced Model Access!** 🚀
This integration bridges the gap between [tgpt](https://github.com/aandrew-me/tgpt) and [Open WebUI](https://github.com/open-webui), allowing you to seamlessly use tgpt's functionalities, including access to models like Phind, directly within your Open WebUI interface.
## ✨ Features
* **Integrate tgpt with Open WebUI:** Use tgpt's command-line power within the familiar Open WebUI environment.
* **Access Phind and more:** Leverage tgpt's model support, including Phind and other models accessible through tgpt, in Open WebUI.
* **Simple Setup:** Designed for users who already have both tgpt and Open WebUI installed.
## 🛠️ Installation
1. **Prerequisites:**
* Ensure you have [tgpt](https://github.com/aandrew-me/tgpt) installed and configured.
* Ensure you have [Open WebUI](https://github.com/open-webui) installed and running.
2. **Copy the Integration Code:**
* Save the provided Python code as a `.py` file (e.g., `tgpt_integration.py`).
* Place this file in the appropriate directory within your Open WebUI installation to register it as a custom function or extension. (Consult Open WebUI documentation for the specific location for custom integrations).
## ⚙️ Configuration
* **TGPT Binary Path:**
* The integration assumes `tgpt` is located at `/usr/local/bin/tgpt`.
* If your `tgpt` binary is in a different location, you will need to modify the `TGPT_PATH` variable in the `Valves` class within the Python code.
```python
class Valves(BaseModel):
TGPT_PATH: str = Field(
default="/usr/local/bin/tgpt", description="Path to the tgpt binary"
)
```
* **To change the path:** Edit the `default` value in the code to the correct path of your `tgpt` executable.
## 🚀 Usage
1. **Select the `TGPT Integration` function** within Open WebUI when choosing a model or function for chat interactions.
2. **Interact as usual:** Type your prompts in the Open WebUI chat interface.
3. **tgpt processes the query:** The integration will send your query to the `tgpt` command-line tool in the backend.
4. **Response in Open WebUI:** The response from `tgpt` will be displayed in your Open WebUI chat window.
## 📜 Code Snippet
```python
"""
title: tgpt
author: Sassanix
author_url: https://github.com/sassanix/tgptopnwebuifunction
funding_url: https://github.com/sassanix/tgptopnwebuifunction
version: 0.1
"""
import subprocess
from pydantic import BaseModel, Field
from typing import Union, Generator, Iterator
class Pipe:
class Valves(BaseModel):
TGPT_PATH: str = Field(
default="/usr/local/bin/tgpt", description="Path to the tgpt binary"
)
def __init__(self):
self.type = "pipe"
self.id = "tgpt_integration"
self.name = "TGPT Integration"
self.valves = self.Valves()
def pipe(self, body: dict) -> Union[str, Generator, Iterator]:
query = body.get("messages", [{}])[-1].get("content", "")
if not query:
return "No query provided."
try:
# Use subprocess to call tgpt with the query
result = subprocess.run(
[self.valves.TGPT_PATH, "--quiet", query],
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
)
if result.returncode != 0:
return f"Error executing tgpt: {result.stderr.strip()}"
return result.stdout.strip()
except Exception as e:
return f"Error: {str(e)}"
```
## 👨💻 Author
* **Sassanix**
* [GitHub](https://github.com/sassanix)
* [Project Repository](https://github.com/sassanix/tgptopnwebuifunction)
## ⚖️ License
* [Specify License here, e.g., MIT License] (Add your project's license information here)
---
**Enjoy using tgpt within Open WebUI!** 🎉
# 🔌 tgpt-open-webui-integration
**Connect tgpt to Open WebUI for Enhanced Model Access!** 🚀
Integrates [tgpt](https://github.com/aandrew-me/tgpt) with [Open WebUI](https://github.com/open-webui) to use tgpt models like Phind within Open WebUI. See README for installation, configuration, and usage.