A locally run Web UI for GPT 3.5 aka ChatGPT's API
Summarization is a technique for keeping track of conversations that exceed the 4096 token limit. The backend will keep track of a running summary, and any time messages are too old to be included in the next API request, a seperate API request will summarize those messages and add it to the running summary.
Context/Memories is another technique for keeping track of conversations that exceed the 4096 token limit. The backend will keep track of a list of context memory groups, where each time a group of messages are too old to be included in the next API request, a seperate API request will generate a context and list of memories from those messages. These memories have keywords so future AI requests can be preceded with an API call to select the most relevant keywords, and the memories associated with those keywords will be added to the AI request.
Run npm install
from the base directory (in terminal, git bash, powershell, etc. Lookup a tutorial for node if you're not familiar with it)
You may need to install Angular CLI version 15.2.1 seperately. If you have issues with the following step.
Follow the guides below to start the frontend and backend servers.
Run npm run startBackend
for the backend server. The application will not reload until you close the server and restart it.
Run npm run startFrontend
for the frontend server. Navigate to http://localhost:4200/
. The application will automatically reload if you change any of the source files.
This project was generated with Angular CLI version 15.2.1.
Run ng generate component component-name
to generate a new component. You can also use ng generate directive|pipe|service|class|guard|interface|enum|module
.
Run ng build
to build the project. The build artifacts will be stored in the dist/
directory.
To get more help on the Angular CLI use ng help
or go check out the Angular CLI Overview and Command Reference page.