2023-03-03 00:28:46 +00:00
# AIChat
2023-03-03 00:33:57 +00:00
[![CI ](https://github.com/sigoden/aichat/actions/workflows/ci.yaml/badge.svg )](https://github.com/sigoden/aichat/actions/workflows/ci.yaml)
[![Crates ](https://img.shields.io/crates/v/aichat.svg )](https://crates.io/crates/aichat)
2024-03-11 10:48:54 +00:00
All-in-one CLI tool for 10+ AI platforms, including OpenAI, Gemini, Claude, Mistral, LocalAI, Ollama, VertexAI, Ernie, Qianwen...
2023-03-06 02:33:13 +00:00
2024-03-06 01:32:00 +00:00
Command Mode:
2023-03-21 01:45:48 +00:00
2024-03-06 01:32:00 +00:00
![command mode ](https://github.com/sigoden/aichat/assets/4012553/2ab27e1b-4078-4ea3-a98f-591b36491685 )
2023-03-21 01:45:48 +00:00
2024-03-06 01:32:00 +00:00
Chat REPL mode:
2023-03-21 01:45:48 +00:00
2024-03-06 01:32:00 +00:00
![chat-repl mode ](https://github.com/sigoden/aichat/assets/4012553/13427d54-efd5-4f4c-b17b-409edd30dfa3 )
2023-03-03 00:33:57 +00:00
2023-03-03 00:28:46 +00:00
## Install
2023-11-28 03:58:28 +00:00
### Use a package management tool
2023-03-03 00:28:46 +00:00
2023-11-27 10:02:38 +00:00
For Rust programmer
```sh
cargo install aichat
2023-03-03 00:28:46 +00:00
```
2023-11-27 10:02:38 +00:00
For macOS Homebrew or a Linuxbrew user
```sh
brew install aichat
```
For Windows Scoop user
```sh
scoop install aichat
```
2024-03-11 10:48:54 +00:00
For Android Termux user
2023-11-27 10:02:38 +00:00
```sh
pkg install aichat
2023-03-03 00:28:46 +00:00
```
2024-03-11 10:48:54 +00:00
### Binaries for macOS, Linux, and Windows
2023-03-03 00:28:46 +00:00
2024-03-11 10:48:54 +00:00
Download it from [GitHub Releases ](https://github.com/sigoden/aichat/releases ), unzip, and add aichat to your `$PATH` .
2023-03-03 00:28:46 +00:00
2023-03-03 03:44:23 +00:00
## Features
2024-03-02 07:52:33 +00:00
- Support most of the LLM platforms
2024-03-06 01:32:00 +00:00
- OpenAI: GPT3.5/GPT4 (paid, vision)
2024-03-02 07:52:33 +00:00
- Gemini (free, vision)
2024-03-06 01:32:00 +00:00
- Claude: Claude2/Claude3 (paid)
2024-03-02 07:52:33 +00:00
- Mistral (paid)
- LocalAI (free, local, vision)
- Ollama (free, local)
- Azure-OpenAI (paid)
2024-03-11 10:48:54 +00:00
- VertexAI: Gemini-1/Gemini-1.5 (paid, vision)
2024-03-02 07:52:33 +00:00
- Ernie (paid)
- Qianwen (paid, vision)
2024-03-06 01:32:00 +00:00
- Support [Command Mode ](#command ) and [Chat-REPL Mode ](#chat-repl )
- Support [roles ](#roles )
- Support sessions (context-aware conversation)
2023-12-27 01:27:59 +00:00
- Support multimodal models (vision)
2024-02-24 11:13:48 +00:00
- Execute commands using natural language
2024-03-06 01:32:00 +00:00
- Shell integration
2023-12-27 01:27:59 +00:00
- Syntax highlighting for markdown and 200+ languages in code blocks
2023-11-27 10:02:38 +00:00
- Save messages/sessions
2024-03-02 07:52:33 +00:00
- Stream/Non-stream output
2024-03-14 02:31:23 +00:00
- [Custom theme ](https://github.com/sigoden/aichat/wiki/Custom-Theme )
2024-03-02 07:52:33 +00:00
- With proxy
2023-03-03 03:44:23 +00:00
2023-03-03 00:28:46 +00:00
## Config
2023-07-10 03:17:42 +00:00
On first launch, aichat will guide you through the configuration.
2023-03-03 02:34:02 +00:00
```
> No config file, create a new one? Yes
2023-10-31 08:50:08 +00:00
> AI Platform: openai
2024-03-11 10:48:54 +00:00
> API Key: <your_api_key_here>
2023-03-03 02:34:02 +00:00
```
2023-10-31 09:34:16 +00:00
Feel free to adjust the configuration according to your needs.
2023-03-03 00:28:46 +00:00
2023-03-03 01:38:53 +00:00
```yaml
2023-10-31 09:34:16 +00:00
model: openai:gpt-3.5-turbo # LLM model
2024-03-22 11:24:22 +00:00
temperature: 1.0 # LLM temperature
2023-10-31 09:34:16 +00:00
save: true # Whether to save the message
2023-10-27 10:44:27 +00:00
highlight: true # Set false to turn highlight
2023-10-31 09:34:16 +00:00
light_theme: false # Whether to use a light theme
2023-10-31 10:34:38 +00:00
wrap: no # Specify the text-wrapping mode (no, auto, < max-width > )
2023-10-30 02:07:01 +00:00
wrap_code: false # Whether wrap code block
2023-10-27 10:44:27 +00:00
auto_copy: false # Automatically copy the last output to the clipboard
2023-10-31 09:34:16 +00:00
keybindings: emacs # REPL keybindings. values: emacs, vi
2023-11-07 11:14:58 +00:00
prelude: '' # Set a default role or session (role:< name > , session:< name > )
2024-03-04 03:08:59 +00:00
compress_threshold: 1000 # Compress session if tokens exceed this value (valid when >=1000)
2023-10-27 10:44:27 +00:00
2023-10-31 08:50:08 +00:00
clients:
- type: openai
api_key: sk-xxx
2023-10-29 02:44:30 +00:00
2023-11-03 01:56:15 +00:00
- type: localai
2023-10-31 08:50:08 +00:00
api_base: http://localhost:8080/v1
2023-11-03 01:56:15 +00:00
models:
- name: gpt4all-j
2024-03-06 00:35:40 +00:00
max_input_tokens: 8192
2023-03-03 00:28:46 +00:00
```
2023-11-27 10:02:38 +00:00
Take a look at the [config.example.yaml ](config.example.yaml ) for the complete configuration details.
2023-03-03 00:28:46 +00:00
2023-11-27 10:02:38 +00:00
There are some configurations that can be set through environment variables. For more information, please refer to the [Environment Variables ](https://github.com/sigoden/aichat/wiki/Environment-Variables ) page.
2023-03-08 23:04:12 +00:00
2024-03-06 01:32:00 +00:00
## Command
2023-03-03 00:28:46 +00:00
2024-03-06 01:32:00 +00:00
```
Usage: aichat [OPTIONS] [TEXT]...
2023-03-03 00:28:46 +00:00
2024-03-06 01:32:00 +00:00
Arguments:
[TEXT]... Input text
2023-03-13 02:26:33 +00:00
2024-03-06 01:32:00 +00:00
Options:
-m, --model < MODEL > Choose a LLM model
-r, --role < ROLE > Choose a role
-s, --session [< SESSION > ] Create or reuse a session
-e, --execute Execute commands using natural language
-c, --code Generate only code
-f, --file < FILE > ... Attach files to the message to be sent
-H, --no-highlight Disable syntax highlighting
-S, --no-stream No stream output
-w, --wrap < WRAP > Specify the text-wrapping mode (no, auto, < max-width > )
--light-theme Use light theme
--dry-run Run in dry run mode
--info Print related information
--list-models List all available models
--list-roles List all available roles
--list-sessions List all available sessions
-h, --help Print help
-V, --version Print version
```
2023-03-03 00:28:46 +00:00
2024-03-06 01:32:00 +00:00
Here are some practical examples:
```sh
aichat # Start in REPL mode
aichat -e install nvim # Execute
aichat -c fibonacci in js # Code
aichat -s # REPL + New session
2024-03-07 12:31:06 +00:00
aichat -s session1 # REPL + New/Reuse 'session1'
2024-03-06 01:32:00 +00:00
aichat --info # System info
aichat -r role1 --info # Role info
2024-03-07 12:31:06 +00:00
aichat -s session1 --info # Session info
2024-03-06 01:32:00 +00:00
cat data.toml | aichat -c to json > data.json # Pipe IO
aichat --file a.png b.png -- diff images # Attach files
2023-03-03 00:28:46 +00:00
```
2024-03-06 01:32:00 +00:00
### Execute commands using natural language
Simply input what you want to do in natural language, and aichat will prompt and run the command that achieves your intent.
2023-07-10 03:17:42 +00:00
2023-03-03 00:28:46 +00:00
```
2024-03-06 01:32:00 +00:00
aichat -s < text > ...
```
2023-03-05 03:11:15 +00:00
2024-03-06 01:32:00 +00:00
Aichat is aware of OS and `$SHELL` you are using, it will provide shell command for specific system you have. For instance, if you ask `aichat` to update your system, it will return a command based on your OS. Here's an example using macOS:
```sh
aichat -e update my system
# sudo softwareupdate -i -a
# ? [e]xecute, [d]escribe, [a]bort: (e)
2023-03-08 13:35:21 +00:00
```
2024-03-06 01:32:00 +00:00
The same prompt, when used on Ubuntu, will generate a different suggestion:
```sh
aichat -e update my system
# sudo apt update && sudo apt upgrade -y
# ? [e]xecute, [d]escribe, [a]bort: (e)
```
We can still use pipes to pass input to aichat and generate shell commands:
```sh
aichat -e POST localhost with < data.json
# curl -X POST -H "Content-Type: application/json" -d '{"a": 1, "b": 2}' localhost
# ? [e]xecute, [d]escribe, [a]bort: (e)
```
2023-10-27 11:01:22 +00:00
2024-03-06 01:32:00 +00:00
We can also pipe the output of aichat which will disable interactive mode.
```sh
aichat -e find all json files in current folder | pbcopy
2023-10-27 11:01:22 +00:00
```
2024-03-06 01:32:00 +00:00
### Shell integration
This is a **very handy feature** , which allows you to use `aichat` shell completions directly in your terminal, without the need to type `aichat` with prompt and arguments. This feature puts `aichat` completions directly into terminal buffer (input line), allowing for immediate editing of suggested commands.
![aichat-integration ](https://github.com/sigoden/aichat/assets/4012553/873ebf23-226c-412e-a34f-c5aaa7017524 )
To install shell integration, go to [./scripts/shell-integration ](https://github.com/sigoden/aichat/tree/main/scripts/shell-integration ) to download the script and source the script in rc file. After that restart your shell. You can invoke the completion with `alt+e` hotkey.
## Generate Code
By using the `--code` or `-c` parameter, you can specifically request pure code output, for instance:
```
aichat --code a echo server in node.js
2023-10-27 11:01:22 +00:00
```
2024-03-06 01:32:00 +00:00
```js
const net = require('net');
const server = net.createServer(socket => {
socket.on('data', data => {
socket.write(data);
});
socket.on('end', () => {
console.log('Client disconnected');
});
});
server.listen(3000, () => {
console.log('Server running on port 3000');
});
```
Since it is valid js code, we can redirect the output to a file:
```
aichat --code a echo server in node.js > echo-server.js
node echo-server.js
```
2023-03-05 03:11:15 +00:00
2023-03-06 02:33:13 +00:00
## Chat REPL
aichat has a powerful Chat REPL.
2024-03-06 01:32:00 +00:00
The REPL supports:
2023-07-10 03:17:42 +00:00
2024-03-03 07:25:21 +00:00
- Tab autocomplete
2024-03-06 01:32:00 +00:00
- [Custom REPL Prompt ](https://github.com/sigoden/aichat/wiki/Custom-REPL-Prompt )
- Emacs/Vi keybinding
2023-12-24 08:04:18 +00:00
- Edit/paste multiline text
2024-02-24 11:13:48 +00:00
- Open an editor to modify the current prompt
2023-03-10 06:51:15 +00:00
- Undo support
2023-03-06 02:33:13 +00:00
2023-10-30 05:36:08 +00:00
### `.help` - print help message
2023-03-10 02:27:24 +00:00
```
2023-12-20 23:21:17 +00:00
> .help
2023-10-30 06:17:15 +00:00
.help Print this help message
2023-10-30 11:26:06 +00:00
.info Print system info
2023-10-30 06:17:15 +00:00
.model Switch LLM model
2023-11-07 11:14:58 +00:00
.role Use a role
2023-10-30 11:26:06 +00:00
.info role Show role info
2023-10-30 06:17:15 +00:00
.exit role Leave current role
.session Start a context-aware chat session
2023-10-30 11:26:06 +00:00
.info session Show session info
2024-03-06 01:32:00 +00:00
.clear messages Clear messages in the session
2023-10-30 06:17:15 +00:00
.exit session End the current session
2023-11-27 06:04:50 +00:00
.file Attach files to the message and then submit it
2023-10-30 06:17:15 +00:00
.set Modify the configuration parameters
.copy Copy the last reply to the clipboard
2023-03-10 02:27:24 +00:00
.exit Exit the REPL
2023-11-08 03:19:45 +00:00
Type ::: to begin multi-line editing, type ::: to end it.
2024-02-20 13:31:42 +00:00
Press Ctrl+O to open an editor to modify the current prompt.
2023-03-13 02:26:33 +00:00
Press Ctrl+C to abort readline, Ctrl+D to exit the REPL
2023-10-30 06:17:15 +00:00
2023-03-06 02:33:13 +00:00
```
2023-10-30 11:26:06 +00:00
### `.info` - view information
2023-03-06 05:00:43 +00:00
```
2023-12-20 23:21:17 +00:00
> .info
2023-10-27 10:44:27 +00:00
model openai:gpt-3.5-turbo
2023-10-30 07:08:49 +00:00
temperature -
2023-11-07 23:08:48 +00:00
dry_run false
2023-03-06 05:00:43 +00:00
save true
highlight true
2023-03-13 02:26:33 +00:00
light_theme false
2023-10-30 07:08:49 +00:00
wrap no
wrap_code false
2023-11-07 23:08:48 +00:00
auto_copy false
2023-10-30 07:08:49 +00:00
keybindings emacs
2023-11-07 15:07:42 +00:00
prelude -
2024-03-06 01:32:00 +00:00
compress_threshold 1000
2023-11-02 08:08:11 +00:00
config_file /home/alice/.config/aichat/config.yaml
roles_file /home/alice/.config/aichat/roles.yaml
messages_file /home/alice/.config/aichat/messages.md
sessions_dir /home/alice/.config/aichat/sessions
2023-03-06 05:00:43 +00:00
```
2023-10-30 05:36:08 +00:00
### `.model` - choose a model
2023-03-21 01:45:48 +00:00
```
2023-10-27 10:44:27 +00:00
> .model openai:gpt-4
2024-03-06 01:32:00 +00:00
> .model ollama:llama2
2023-03-21 01:45:48 +00:00
```
2024-03-11 10:46:36 +00:00
> You can easily enter model name using autocomplete.
2023-10-30 05:36:08 +00:00
### `.role` - let the AI play a role
2023-03-08 13:35:21 +00:00
2023-07-10 03:17:42 +00:00
Select a role:
2023-03-10 02:27:24 +00:00
2023-03-08 13:35:21 +00:00
```
2023-12-20 23:21:17 +00:00
> .role emoji
2023-03-10 02:27:24 +00:00
```
2023-10-30 11:30:28 +00:00
Send message with the role:
2023-07-10 03:17:42 +00:00
2023-03-10 02:27:24 +00:00
```
2023-12-20 23:21:17 +00:00
emoji> hello
2023-03-10 02:27:24 +00:00
👋
2023-03-08 13:35:21 +00:00
```
2023-10-30 11:26:06 +00:00
Leave current role:
2023-07-10 03:17:42 +00:00
2023-03-08 13:35:21 +00:00
```
2023-12-20 23:21:17 +00:00
emoji> .exit role
2023-03-10 02:27:24 +00:00
2023-12-20 23:21:17 +00:00
> hello
2023-03-10 02:27:24 +00:00
Hello there! How can I assist you today?
2023-03-08 13:35:21 +00:00
```
2023-10-30 11:26:06 +00:00
Show role info:
```
2023-12-20 23:21:17 +00:00
emoji> .info role
2023-10-30 11:26:06 +00:00
name: emoji
prompt: I want you to translate the sentences I write into emojis. I will write the sentence, and you will express it with emojis. I just want you to express it with emojis. I don't want you to reply with anything but emoji. When I need to tell you something in English, I will do it by wrapping it in curly brackets like {like this}.
temperature: null
```
2023-11-07 15:07:42 +00:00
Temporarily use a role to send a message.
```
2023-12-20 23:21:17 +00:00
> ::: .role emoji
2023-11-07 15:07:42 +00:00
hello world
2023-11-08 03:19:45 +00:00
:::
2023-11-07 15:07:42 +00:00
👋🌍
2023-12-20 23:21:17 +00:00
>
2023-11-07 15:07:42 +00:00
```
2023-10-30 11:30:28 +00:00
### `.session` - context-aware conversation
2023-03-06 02:33:13 +00:00
2023-03-10 02:27:24 +00:00
By default, aichat behaves in a one-off request/response manner.
2023-10-28 14:22:37 +00:00
You should run aichat with `-s/--session` or use the `.session` command to start a session.
2023-03-06 02:33:13 +00:00
2023-03-08 13:35:21 +00:00
```
2023-12-20 23:21:17 +00:00
> .session
2023-11-01 14:15:55 +00:00
2023-12-20 23:21:17 +00:00
temp) 1 to 5, odd only 0
2023-10-28 13:39:17 +00:00
1, 3, 5
2023-03-10 02:27:24 +00:00
2023-12-20 23:21:17 +00:00
temp) to 7 19(0.46%)
2023-10-28 13:39:17 +00:00
1, 3, 5, 7
2023-03-10 02:27:24 +00:00
2023-12-20 23:21:17 +00:00
temp) .exit session 42(1.03%)
2023-11-01 14:15:55 +00:00
? Save session? (y/N)
2023-03-21 01:45:48 +00:00
```
2023-11-01 14:15:55 +00:00
The prompt on the right side is about the current usage of tokens and the proportion of tokens used,
compared to the maximum number of tokens allowed by the model.
2023-03-21 01:45:48 +00:00
2023-11-27 06:04:50 +00:00
### `.file` - attach files to the message
```
Usage: .file < file > ... [-- text...]
.file message.txt
.file config.yaml -- convert to toml
.file a.jpg b.jpg -- What’ s in these images?
.file https://ibb.co/a.png https://ibb.co/b.png -- what is the difference?
```
2023-11-27 10:02:38 +00:00
> Only the current model that supports vision can process images submitted through `.file` command.
2023-10-30 06:17:15 +00:00
### `.set` - modify the configuration temporarily
```
2024-03-11 10:48:54 +00:00
.set temperature 1.2
.set compress_threshold 1000
.set dry_run true
.set highlight false
.set save false
.set auto_copy true
2023-10-30 09:09:53 +00:00
```
2024-03-06 01:32:00 +00:00
### Roles
2023-10-30 09:09:53 +00:00
2024-03-06 01:32:00 +00:00
We can define a batch of roles in `roles.yaml` .
2023-10-30 09:09:53 +00:00
2024-03-06 01:32:00 +00:00
> Retrieve the location of `roles.yaml` through the REPL `.info` command or CLI `--info` option.
2023-10-30 09:09:53 +00:00
2024-03-06 01:32:00 +00:00
For example, we can define a role:
2024-02-23 05:15:18 +00:00
2024-03-06 01:32:00 +00:00
```yaml
- name: shell
prompt: >
I want you to act as a Linux shell expert.
I want you to answer only with bash code.
Do not provide explanations.
2024-02-23 05:15:18 +00:00
```
2024-03-06 01:32:00 +00:00
Let ChatGPT answer questions in the role of a Linux shell expert.
2024-02-23 05:15:18 +00:00
```
2024-03-06 01:32:00 +00:00
> .role shell
2024-02-23 05:15:18 +00:00
2024-03-06 01:32:00 +00:00
shell> extract encrypted zipfile app.zip to /tmp/app
mkdir /tmp/app
unzip -P PASSWORD app.zip -d /tmp/app
2024-02-23 05:15:18 +00:00
```
2024-03-06 01:32:00 +00:00
AIChat with roles will be a universal tool.
2024-02-23 05:15:18 +00:00
```
2024-03-06 01:32:00 +00:00
$ aichat --role shell extract encrypted zipfile app.zip to /tmp/app
unzip -P password app.zip -d /tmp/app
2024-02-23 05:15:18 +00:00
2024-03-06 01:32:00 +00:00
$ cat README.md | aichat --role spellcheck
2024-02-23 05:15:18 +00:00
```
2024-03-06 01:32:00 +00:00
For more details about roles, please visit [Role Guide ](https://github.com/sigoden/aichat/wiki/Role-Guide ).
2024-02-26 03:28:25 +00:00
2023-03-03 00:28:46 +00:00
## License
2024-02-24 11:13:48 +00:00
Copyright (c) 2023-2024 aichat-developers.
2023-03-03 00:28:46 +00:00
2024-02-23 05:15:18 +00:00
Aichat is made available under the terms of either the MIT License or the Apache License 2.0, at your option.
2023-03-03 00:28:46 +00:00
2023-03-04 13:43:29 +00:00
See the LICENSE-APACHE and LICENSE-MIT files for license details.