https://github.com/mvrilo/goprivategpt

privategpt implemented with Go
https://github.com/mvrilo/goprivategpt

go golang gpt llama llm privategpt sqlite weaviate

Last synced: 3 months ago
JSON representation

privategpt implemented with Go

• Host: GitHub
• URL: https://github.com/mvrilo/goprivategpt
• Owner: mvrilo
• Created: 2023-06-22T14:47:38.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-03-18T01:37:42.000Z (10 months ago)
• Last Synced: 2024-10-18T23:14:52.671Z (3 months ago)
• Topics: go, golang, gpt, llama, llm, privategpt, sqlite, weaviate
• Language: Go
• Homepage:
• Size: 66.4 KB
• Stars: 10
• Watchers: 1
• Forks: 2
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# goprivategpt

Implementation of PrivateGPT in Go using [langchaingo](https://github.com/tmc/langchaingo) and [go-llama.cpp](https://github.com/go-skynet/go-llama.cpp).

Supported file extensions:

- `csv`

- `html`

- `txt`

Supported vector stores:

- [sqlite-vss](https://github.com/asg017/sqlite-vss)

- [weaviate](https://github.com/weaviate/weaviate)

Requirements for running:

- `ggml model`

### Dependencies

- `brew install libomp`

### Building

- Mac:

```

make build-metal

```

- Linux:

```

make build

```

### Usage

```

A way for you to interact with your documents.

Usage:

  goprivategpt [command]

Available Commands:

  ask         completes a given input

  help        Help about any command

  ingest      Ingests documents from source directory into the vector store

  server      Starts the http server

Flags:

  -h, --help               help for goprivategpt

  -s, --storeaddr string   Vector store filename or address (default "goprivategpt.db")

  -t, --threads int        Number of threads for LLM (default 8)

  -n, --tokens int         Number of max tokens in response (default 512)

Use "goprivategpt [command] --help" for more information about a command.

```