Skip to content
BranchesTags
 
 

Repository files navigation

AmigaGPT

AmigaGPT is a versatile multi-provider AI client for AmigaOS 3.x, 4.1 and MorphOS. This powerful tool brings the capabilities of leading AI providers — OpenAI, Google Gemini, xAI Grok, Anthropic Claude, and custom LLM servers — to your Amiga system, enabling text generation, question answering, and creative exploration. AmigaGPT can also generate stunning images and includes support for speech output, making it easier than ever to interact with AI on your Amiga. Designed to integrate seamlessly with your system, AmigaGPT delivers modern AI technology while embracing the timeless Amiga experience.

Screenshot 2023-06-15 at 10 26 38 pm

Features

  • State-of-the-art language and image models

AmigaGPT supports multiple AI providers for chat: OpenAI (o1, o3, o4, GPT-5, GPT-4o, GPT-4 and GPT-3.5), Google Gemini, xAI Grok, Anthropic Claude, and custom LLM servers compatible with the OpenAI API. Switch between providers and models from the Chat Provider Settings.

  • AI Image Generation

AmigaGPT can generate images from a prompt using OpenAI (DALL-E 2, DALL-E 3, GPT-Image-1), Google Gemini, and xAI Grok. You can view and save the images right inside the app, with a choice of output format (PNG or JPEG).

  • Seamless integration with AmigaOS and MorphOS

AmigaGPT takes full advantage of the MUI framework to provide a smooth, native user experience that is responsive and easy to use.

  • UI customisation

You can customise the look and feel of the application, including the ability to choose the fonts, colours and a choice of opening in the Workbench screen or a custom screen.

  • Speech capability

AmigaGPT supports multiple speech providers, configurable from Speech Provider Settings. For cloud-based voices, it supports OpenAI TTS (high quality 16-bit voices), xAI Grok TTS, ElevenLabs, and OpenVox (requires AHI). For offline synthesis, AmigaOS 3 can use the Amiga's built-in narrator.device (v34 and v37) with translator.library, and AmigaOS 4.1 supports flite.device. Speech runs in the background so the app stays responsive, with a dedicated Stop button to interrupt playback.

  • Code Interpreter

AmigaGPT can let the AI write and run Python in a provider-hosted sandbox for calculations, data analysis, and generated files. Enable it in Chat Provider Settings for OpenAI, Google Gemini, xAI Grok, Anthropic Claude, and custom profiles that use the Responses, Messages, or Gemini generateContent endpoints. It is hidden for custom chat/completions servers that do not offer a hosted interpreter. You can also use the CI=CODEINTERPRETER flag in the ARexx SENDMESSAGE command.

  • Shell Tool

AmigaGPT can let the AI execute AmigaDOS commands directly on your Amiga. Enable it in Chat Provider Settings and ask the AI to list files, create directories, show file contents, and more. All commands require your approval before execution for security.

  • Web Search

AmigaGPT can instruct the AI to search the web for up-to-date information. Enable or disable it in Chat Provider Settings, or use the W=WEBSEARCH flag in the ARexx SENDMESSAGE command.

  • ARexx integration

AmigaGPT includes ARexx support, allowing you to control the application programmatically from external scripts or other applications. You can send prompts, generate images, and utilize speech synthesis through simple ARexx commands.

System Requirements

Ensure you have the necessary system requirements:

Installation

  • For AmigaOS 3 & 4, Install AmiSSL and a TCP/IP stack if not already done so
  • Download and install MUI. Version 5 recommended, version 3 minimum. Reboot.
  • Download and install codesets.library, render.library, MCC_Guigfx, MCC_NList and MCC_TextEditor
  • Download the latest release of AmigaGPT
  • Extract the amigagpt.lha archive to a temporary location
  • Run the provided installer

Optional steps to enable speech functionality

AmigaGPT supports reading the output aloud. How AmigaGPT does this depends on your chosen speech provider. Cloud-based voices (OpenAI TTS, xAI Grok TTS, ElevenLabs, OpenVox) work on every supported OS and require AHI. Offline synthesis uses OS-native devices.

Installing AHI for cloud-based voices

If your OS does not come with AHI installed, you can get it from https://aminet.net/package/driver/audio/ahiusr_4.18

AmigaOS 3

AmigaGPT supports reading the output aloud. This requires a file called narrator.device which cannot be included with AmigaGPT because it is still under copyright. Therefore, you must copy this file legally from your Workbench disks so that AmigaGPT will be able to synthesise speech. There are 2 versions of narrator.device supported, v34 and v37.

v34 is the original version that came with Workbench 1.x. v37 was an updated version included with Workbench 2.0.x. It has more features and sounds more natural, however it does sound quite different which is why AmigaGPT supports you installing both versions and your choice of version to be used can be selected in the Speech menu in the app.

Regardless of which version of narrator.device you choose to install (or both), AmigaGPT requires the free third party translator.library v42 or higher. This works with both versions of narrator.device.

Installing translator.library

v42 is all AmigaGPT requires, and it is available as a standalone install:

  • Download http://aminet.net/util/libs/translator42.lha and extract the archive to any convenient location on your Amiga such as RAM:
  • Navigate to that directory and double click the Install program
  • Run the installer using all the default settings
  • Reboot your Amiga - It will not work until the system is restarted

Optional: patching to v43

v43 is a later revision with improved pronunciation. AmigaGPT does not need it - earlier documentation said it was required, but that is no longer the case - and speech works fine on v42. If you want it anyway, install v42 as above and then apply the patch:

  • Download http://aminet.net/util/libs/Tran43pch.lha and extract it to a location of your choice
  • Navigate to that directory and double click the Install program
  • Run the installer using all the default settings
  • Reboot your Amiga

Installing narrator.device v34

  • Insert your Workbench 1.x disk and copy df0:devs/narrator.device to AMIGAGPT:devs/speech/34

Installing narrator.device v37

  • Insert your Workbench 2.0.x (you cannot use 2.1 because the speech libraries were removed after version 2.0.4) disk and copy df0:devs/narrator.device to AMIGAGPT:devs/speech/37

AmigaOS 4

MorphOS

MorphOS has no offline speech system. Use one of the cloud voices (OpenAI, xAI, ElevenLabs or OpenVox) with AHI installed.

Launching AmigaGPT

  • Launch the application by double-clicking the AmigaGPT icon
  • You may also launch the app in the command line but before you do, run the command STACK 32768 to give the program 32kb of stack since the default stack size for apps launched from the shell is 4kb and this is not enough for AmigaGPT and will cause random crashes due to stack overflow. This is not required when you launch the app by double clicking the icon since the stack size is saved in the icon

Usage

When launching for the first time you will need to enter your AI provider's API key before you can start chatting. Open Chat Provider Settings from the Chat menu to select your provider and enter your key. For OpenAI, navigate to https://platform.openai.com/account/api-keys to generate an API key.

There are 2 main modes of operation: Chat and Image Generation. You can switch between them via the tabs in the top left corner.

Chat

When the app has opened, you are presented with a text input box. You can type any prompt into this box and press "Send" to see the AI's response. The generated text appears in the box above the input. You can choose to have this text read aloud using the "Speech" menu option; configure the speech provider and voice in "Speech Provider Settings" from the Speech menu. To enable or disable web search or the code interpreter, open Chat Provider Settings from the Chat menu. To choose the chat model and provider, use the same window.

Use Attach files... to select one or more files to send with the next message, or pass them with the A=ATTACH argument on the ARexx SENDMESSAGE command. The selected files are shown beside the chat controls and can be removed with Clear before sending. OpenAI, xAI Grok, Anthropic Claude and Google Gemini upload the file first and then reference it, so the chat request stays small on classic Amiga TCP stacks. Other providers receive images and documents in their native multimodal request format. The type of each file is determined by looking inside it with libmagic -- the same code the Unix file command uses -- rather than from its name, so extensionless Amiga files such as User-Startup are sent as plain text and a file with a misleading name is sent as what it actually is. This uses the magic.mgc database that the installer copies into the AmigaGPT drawer by default, with C: offered as an alternative so one copy can be shared with the file package from Aminet; AmigaGPT checks PROGDIR:magic.mgc first, then AMIGAGPT:magic.mgc, then C:magic.mgc, and falls back to guessing from the file extension if none of them is present. ZIP archives are sent as application/zip. On OpenAI they are placed in the Code Interpreter container (enable Code Interpreter so the model can unpack them); on Anthropic Claude they are uploaded to the code-execution container the same way. To keep memory use practical on classic Amigas, each file is limited to 8 MB and all files in one request are limited to 16 MB.

When an LLM response includes downloadable files, they are listed beneath the assistant message and Save received files... becomes available. Choose a destination drawer and AmigaGPT will save every returned file there without overwriting an existing file; numeric suffixes are added when necessary. From ARexx, SENDMESSAGE downloads them to D=DESTINATION if you give a drawer, or to the current directory if you omit it. Provider and model support for particular input types and returned files varies.

To the left of the chat box is a conversation list which you can use to go to another saved conversation. New conversations can be created with the "New chat" button and conversations can be removed with the "Delete chat" button.

You can also chat to AmigaGPT within the Shell as long as either AmigaGPT or the AmigaGPTD daemon are running. Just type askgpt into the shell and it will ask you what you want to ask. Or just ask right away with the command askgpt What is the capital of Australia?.

askgpt [GUI] [NEWCHAT] [PROFILE=name] question
  • GUI opens a small window for the question and the answer instead of printing to the Shell. This needs RxMUI and rmh.library; without them AskGPT falls back to a simple requester
  • NEWCHAT clears the conversation history first, so the question is not answered in the context of whatever came before
  • PROFILE=name (or PR=name) uses a particular chat profile instead of the active one

Because the daemon keeps its history between calls, successive askgpt commands continue the same conversation until you use NEWCHAT or reboot.

Image Generation

To generate images, open "Image Provider Settings" from the "Image" menu to choose your provider (OpenAI, Google Gemini, or xAI Grok) and model, then type your prompt in the text box and hit the "Create Image" button. You can also select the output format (PNG or JPEG) from the Image Provider Settings. When the image has been downloaded to your Amiga, you are then able to open it to your desired scale, or save a copy to a new location. Note that AmigaGPT will automatically save all generated images until you delete them.

Chat and Image Provider Settings

AmigaGPT supports multiple AI providers. Open "Chat Provider Settings" from the "Chat" menu or "Image Provider Settings" from the "Image" menu.

The provider list includes built-in entries for OpenAI, Google Gemini, xAI Grok and Anthropic Claude (chat only); these cannot be removed. For built-in providers you can set the API key and choose the model; host, port and other connection settings are fixed. Known models are shown for each built-in; for OpenAI you can also use "Fetch Models" to refresh the list. You can add custom provider profiles (e.g. LM Studio or Ollama) with New, which creates a blank profile you have full control over, or copy an existing one with Duplicate. Test sends a small request using the settings currently on screen and reports PASS or FAIL together with the raw server response. Custom profile names cannot match built-in provider names.

When you switch to another profile without saving, AmigaGPT will ask whether to save your changes (Yes, No or Cancel).

Speech Provider Settings

Open "Speech Provider Settings..." from the "Speech" menu. This is where the speech system is chosen and configured; in earlier versions of AmigaGPT this was done from the Speech menu itself.

Like the chat and image windows it has a Profiles list with New, Duplicate and Delete, a Profile Name field, and a Test button that speaks a sample using the settings currently on screen. The Speech system cycle chooses between:

Speech system Platform Needs
Workbench 1.x v34 AmigaOS 3 narrator.device v34 + translator.library v42+
Workbench 2.0 v37 AmigaOS 3 narrator.device v37 + translator.library v42+
Flite AmigaOS 4 flite.device
OpenAI any AHI + OpenAI API key
xAI any AHI + xAI API key
ElevenLabs any AHI + ElevenLabs API key
OpenVox any AHI + an OpenVox server (defaults to 127.0.0.1:8666)

MorphOS has no offline speech system, so use one of the cloud voices.

Depending on the selected system the window shows:

  • Connection settings (cloud systems) - Host, Port, Encryption, Authorization and API Endpoint URL
  • Workbench narrator.device (v34/v37) - Accent (default american.accent), Rate (wpm) (default 150), Pitch (Hz) (default 110), Mode (Natural/Robotic) and Sex (Male/Female). v34 and v37 each keep their own copy of these.
  • Flite Voice - kal, kal16, awb, rms or slt
  • OpenAI - API key, Model (tts-1, tts-1-hd, gpt-4o-mini-tts) with Fetch Models, Voice (alloy, ash, ballad, coral, echo, fable, onyx, nova, sage, shimmer, verse) and OpenAI Voice Instructions
  • xAI - API key, Voice (ara, eve, leo, rex, sal) and Auto-insert speech tags into LLM replies
  • ElevenLabs - API key, Model (Eleven v3, Eleven Multilingual v2, Eleven Flash v2.5, Eleven Turbo v2.5) with Fetch Models, and Voice with Search and Get All Voices
  • OpenVox - models and voices are fetched from your own server

If a profile needs something that is not installed, AmigaGPT warns you which component is missing.

General

The "Project" menu includes an "About" option, which displays information about the program.

In the "Edit" menu, you'll find basic text editing commands: Cut (Right-Amiga + X), Copy (+ C), Paste (+ V), Clear (+ L) and Select all (+ A). They act on whichever text field has the keyboard focus and are greyed out when neither the chat input nor the image prompt is active.

The "Project" menu also includes a Print option (Right-Amiga + P) to print the current conversation.

The "Help" menu has Open Documentation..., which opens AMIGAGPT:AmigaGPT.guide - the full AmigaGuide manual, including a troubleshooting section and a complete menu reference.

The "View" menu allows you to change the appearance of the app, including toggling Fixed Width Fonts and setting the User and Assistant text alignment. MUI Settings... opens the full MUI preferences for AmigaGPT, and Clear MUI Settings resets fonts, colours, screen and window positions back to their defaults.

AmigaGPT can run on its own screen or in Workbench. To configure this, open the MUI Settings menu item from the View menu and configure it in the Screen or System menu in the MUI Settings panel.

If you open in a new screen you have the ability to create a screen for the app to open in. AmigaGPT supports anything from 320x200 all the way up to 4k resolution if using a video card for RTG. Bear in mind text will appear very tiny in resolutions above 1080p so you may want to increase the font size settings from the MUI settings in the View menu when the app opens.

The "Connection" menu allows you to connect via a proxy server. It supports both HTTP and HTTPS proxy servers but if you use an unencrypted HTTP proxy server you can improve the performance of AmigaGPT by removing the need for the encryption of the OpenAI traffic to be done on the system running AmigaGPT. For an easy proxy server you can run on your local network you can try out https://mitmproxy.org

ARexx Support

AmigaGPT now includes ARexx support, allowing you to control the application from external ARexx scripts or other applications. This powerful feature enables seamless integration with your Amiga workflow and automation of repetitive tasks.

Note: MorphOS does not yet have a complete ARexx implementation. You must install the 68k version of ARexx on MorphOS to be able to use ARexx with AmigaGPT.

AmigaGPT also comes with a companion daemon app called AmigaGPTD. You may run this in the shell. The daemon will respond to the same ARexx commands as the main app so you do not need to launch the main AmigaGPT app.

The daemon maintains conversation history in T: so it remembers context across multiple ARexx calls until the next reboot. This allows for follow-up questions and multi-turn conversations. Use the NEWCHAT ARexx command to clear the conversation history and start fresh.

Please note: AmigaGPT and AmigaGPTD listen on 2 different ARexx ports so adjust your scripts accordingly. AmigaGPT listens on AMIGAGPT and AmigaGPTD listens on AMIGAGPTD.

The following ARexx commands are available:

SENDMESSAGE

Sends a message using the selected chat profile and returns the response.

SENDMESSAGE PR=PROFILE/K,M=MODEL/K,S=SYSTEM/K,SF=SYSTEMFILE/K,H=HOST/K,PO=PORT/N,S=SSL/S,K=APIKEY/K,U=USEPROXY/S,PH=PROXYHOST/K,PP=PROXYPORT/N,PS=PROXYUSESSSL/S,PA=PROXYREQUIRESAUTH/S,PU=PROXYUSERNAME/K,PP=PROXYPASSWORD/K,W=WEBSEARCH/S,CI=CODEINTERPRETER/S,A=ATTACH/K,D=DESTINATION/K,P=PROMPT/F

Note: a few short aliases are shared by two arguments in the same template (S for both SYSTEM and SSL, PP for both PROXYPORT and PROXYPASSWORD), and the ? help output prints P=PORT where the command expects PO=PORT. Spell SYSTEM, SSL, PORT, PROXYPORT and PROXYPASSWORD out in full and there is no ambiguity.

  • PR=PROFILE - Optional, the chat profile to use. This can be a built-in profile such as OpenAI, Google Gemini, xAI Grok or Anthropic Claude, or a saved custom chat profile name. Use LISTPROFILES to see available profiles. If omitted, the active chat profile from AmigaGPT config is used
  • M=MODEL - Optional, the chat model to use. Use LISTMODELS PR=<profile> to inspect the models for a profile. If omitted, the selected profile's model is used
  • S=SYSTEM - Optional, system message to include
  • SF=SYSTEMFILE - Optional, file containing the system message to include. If both SF and S are provided, the effective system message is the file contents concatenated with S.
  • H=HOST - Optional, connect to this server instead of the profile's host
  • PORT - Optional, the port to connect to
  • SSL - Optional, use an encrypted connection
  • K=APIKEY - Optional, API key override. If omitted, the selected profile's API key is used
  • U=USEPROXY - Optional, route this request through a proxy server
  • PH=PROXYHOST - Optional, the proxy server address
  • PROXYPORT - Optional, the proxy server port
  • PS=PROXYUSESSSL - Optional, the proxy connection is encrypted
  • PA=PROXYREQUIRESAUTH - Optional, the proxy requires authentication
  • PU=PROXYUSERNAME - Optional, the proxy username
  • PROXYPASSWORD - Optional, the proxy password
  • W=WEBSEARCH - Optional, enable web search. If omitted, the selected profile's setting is used. If provided, it forces web search on even when the selected profile has it disabled
  • CI=CODEINTERPRETER - Optional, enable the hosted code interpreter. If omitted, the selected profile's setting is used. If provided, it forces the code interpreter on even when the selected profile has it disabled
  • A=ATTACH - Optional, one file to send with the prompt, or several paths separated by commas. The files are typed the same way as Attach files... in the GUI. Either ATTACH or PROMPT must be provided
  • D=DESTINATION - Optional, a drawer to save files the model returns. If omitted, they are saved in the current directory. Existing names are not overwritten; a number is added when necessary. When files are saved, RESULT is the reply text followed by a FILES: line and one path per line
  • P=PROMPT - Optional if ATTACH is provided, otherwise required. The prompt or question to send

CREATEIMAGE

Generates an image using the specified model.

CREATEIMAGE PR=PROFILE/K,M=MODEL/K,S=SIZE/K,K=APIKEY/K,D=DESTINATION/K,P=PROMPT/F
  • PR=PROFILE - Optional, the image profile to use. This can be a built-in profile such as OpenAI, Google Gemini or xAI Grok, or a saved custom image profile name. Use LISTPROFILES to see available profiles. If omitted, the active image profile from AmigaGPT config is used
  • M=MODEL - Optional, the image model to use. Use LISTMODELS PR=<profile> to inspect the models for a profile. If omitted, the selected profile's model is used
  • S=SIZE - Optional, image size (use LISTIMAGESIZES to see available sizes). Default is 1024x1024. Note: some providers may ignore unsupported parameters (e.g. xAI currently does not support size) - see xAI Image Generations and Gemini OpenAI compatibility
  • K=APIKEY - Optional, API key override. If omitted, the selected profile's API key is used
  • D=DESTINATION - Optional, the path where the image will be saved. Default is the creation of a temporary file. The destination will be the returned string from this function
  • P=PROMPT - Required, description of the image to generate

SPEAKTEXT

Uses text-to-speech to speak the specified text.

SPEAKTEXT PR=PROFILE/K,M=MODEL/K,V=VOICE/K,I=INSTRUCTIONS/K,K=APIKEY/K,O=OUTPUT/K,F=FORMAT/K,P=PROMPT/F
  • PR=PROFILE - Optional, the speech profile to use. This can be a built-in speech system name or a saved custom speech profile name. Use LISTPROFILES to see available profiles. If omitted, the active speech profile from AmigaGPT config is used
  • M=MODEL - Optional, the OpenAI voice model to use. Use LISTMODELS PR=<profile> to inspect the models for a speech profile. If omitted, the selected speech profile's model is used
  • V=VOICE - Optional voice override (use LISTVOICES to see available voices). For OpenAI speech profiles, this selects the OpenAI voice. For Workbench speech profiles, male, female, male robot, and female robot override the profile's narrator sex/mode. If omitted, the selected speech profile's voice settings are used
  • I=INSTRUCTIONS - Optional, OpenAI voice instructions override
  • K=APIKEY - Optional, API key override. If omitted, the selected speech profile's API key is used
  • O=OUTPUT - Optional, file to write audio data to instead of playing it
  • F=FORMAT - Optional, format of output file (use LISTAUDIOFORMATS to see available formats). Default is MP3 and ignored if OUTPUT is not provided
  • P=PROMPT - Required, the text to speak

NEWCHAT

Clears the daemon's conversation history and starts a new chat. This is useful when you want to start a fresh conversation without context from previous messages.

NEWCHAT

This command takes no parameters and returns "Conversation history cleared" on success.

List Commands

LISTMODELS

Fetches the models available from the server for a profile. If PR is omitted, it fetches model sections for the active chat, image and speech profiles.

LISTMODELS PR=PROFILE/K
  • PR=PROFILE - Optional, the profile to inspect. The same name can match chat, image and speech profiles, and LISTMODELS will include a section for each matching subsystem. Built-in aliases such as ChatGPT, Gemini, Grok and Claude are also accepted
  • LISTMODELS queries the profile's configured server where possible, so the results reflect what that server currently exposes rather than only AmigaGPT's built-in known model lists

LISTAUDIOFORMATS

Lists all available audio formats for saving audio files

LISTAUDIOFORMATS

LISTPROFILES

Lists all available chat, image and speech profiles.

LISTPROFILES

LISTIMAGESIZES

Lists all available image sizes

LISTIMAGESIZES

LISTVOICES

Lists the available TTS voices for a speech profile. If PR is omitted, it uses the active speech profile from AmigaGPT config.

LISTVOICES PR=PROFILE/K
  • PR=PROFILE - Optional, the speech profile to inspect. If the resolved profile uses ElevenLabs, LISTVOICES fetches the available voices from ElevenLabs. For Workbench speech profiles, it returns male, female, male robot, and female robot

Help (?)

Displays a list of available commands

?

Example ARexx Script

/* Simple ARexx script for AmigaGPT */
OPTIONS RESULTS
ADDRESS 'AMIGAGPT'

/* Send a message to GPT and get a response */
'SENDMESSAGE P=What is the capital of France?'
SAY RESULT

/* Generate an image */
'CREATEIMAGE P=A beautiful Amiga computer on a desk'
SAY 'Image saved to:' RESULT

/* Speak some text */
'SPEAKTEXT P=Hello from ARexx!'

EXIT

Bundled ARexx scripts

AmigaGPT installs a set of example scripts into AMIGAGPT:rexx:

Script What it does
askgpt.rexx The Shell front end. Argument parsing, an optional RxMUI interface, profile selection and NEWCHAT
help.rexx Prints the command set, the profiles, the models for the active profiles and the voices for the active speech profile
joke.rexx Asks for a joke and shows it in an RxMUI window, falling back to a plain requester
say.rexx Asks you what to say and speaks it
say_os_version.rexx Runs Version and speaks the result - shows how to feed an AmigaDOS command into AmigaGPT
amiga_image.rexx Generates a picture and opens it in MultiView
textedit_translate_no.rexx Translates the selected text in TextEdit and puts the translation back
show_received_file.rexx Asks the hosted code interpreter for a file, saves it with D=DESTINATION, and opens it in MultiView

The ARexx menu allows you to import and run ARexx scripts. Also, selecting Arexx Shell will open a shell and put all ARexx scripts in trace mode so you can run the scripts line by line for debugging purposes. Select Arexx Shell once again to turn off trace mode and close the shell.

Files and Drawers

Everything AmigaGPT keeps lives under the AMIGAGPT: assign, which the installer adds to S:User-Startup.

Path Contents
AMIGAGPT:AmigaGPT The application
AMIGAGPT:AmigaGPTD The daemon
C:askgpt The Shell command, if you installed it
AMIGAGPT:config.json All settings, profiles and API keys (plain text)
AMIGAGPT:config.bak The previous configuration
AMIGAGPT:chat-history.json Every conversation
AMIGAGPT:image-history.json Every generated image
AMIGAGPT:images/ The generated image files
AMIGAGPT:AmigaGPT.guide This documentation, opened by the Help menu
AMIGAGPT:magic.mgc The file type database (may also be in PROGDIR: or C:)
AMIGAGPT:rexx/ The bundled ARexx scripts
AMIGAGPT:devs/speech/34/ Where narrator.device v34 goes
AMIGAGPT:devs/speech/37/ Where narrator.device v37 goes
LOCALE:Catalogs/<language>/AmigaGPT.catalog The translations
T:AmigaGPTD_history.json The daemon conversation, until the next reboot

If a history file is ever found to be corrupt, AmigaGPT renames it with a .bak extension - or copies it to RAM: if it cannot - tells you what happened, and carries on with an empty history rather than refusing to start.

Language

AmigaGPT follows the language set in your system Locale preferences. English is built into the program; the other 22 languages are installed as catalogs into LOCALE:Catalogs:

Bosnian, Croatian, Czech, Danish, Dutch, English (British), Finnish, French, German, Greek, Hungarian, Italian, Norwegian, Polish, Portuguese, Russian, Serbian, Slovak, Slovenian, Spanish, Swedish and Turkish.

The installer preselects your system language. If no catalog is installed for the current language, AmigaGPT falls back to English. The interface language is separate from the language the AI answers in - to change the latter, put an instruction such as "Always answer in Italian" into the Chat System field of the chat profile.

Models

AmigaGPT supports models from multiple providers. For the latest model information, refer to each provider's documentation:

You can also use the Fetch Models button in Chat Provider Settings to retrieve the current model list directly from the server.

Developing

You can either compile the code natively or with the Docker container.

Native

Building the AmigaOS 3 app

If you would like to build this project from source you will need Bebbo's amiga-gcc toolchain here https://github.com/bebbo/amiga-gcc

Once installed, get the required other SDK's (AmiSSL, Translator, json-c) from https://github.com/sacredbanana/AmigaSDK-gcc and put these in your Amiga dev environment created in the above step.

Building the AmigaOS 4 app

Get this toolchain set up https://github.com/sba1/adtools

Once installed, get the required other SDK's (AmiSSL, Translator, json-c) from https://github.com/sacredbanana/AmigaSDK-gcc and put these in your Amiga dev environment created in the above step.

Docker

You may use pre-prepared Docker images that are able to compile both the AmigaOS 3 and AmigaOS 4 versions of the app.

Just install Docker on your machine and run the build_os3.sh, build_os4.sh or build_morphos.sh scripts depending on which version of the app you want to build. If you want to perform a clean build, you can set the environment variable CLEAN=1 for example you can run CLEAN=1 ./build_os3.sh.

The build app will be saved to the /out directory.

License

AmigaGPT is licensed under the MIT License.

Contributing

We welcome contributions to AmigaGPT! If you have a bug to report, a feature to suggest, or a change you'd like to make to the code, please open a new issue or submit a pull request.

Contributors

Code

Art

  • Mauricio Sandoval - Icon design

Translations

  • Mauricio Sandoval - Spanish
  • Tobias Baeumer - German

Special Thanks

About

AmigaOS 3.1/4.1 and MorphOS application for chatting with ChatGPT or generating images

Resources

Readme
License
Activity
Custom properties

Stars

0 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

Packages

Contributors

Languages