Thox CLI Command Reference
Complete reference guide for all Thox.ai command-line interface commands.
The Thox CLI (thox) is your primary interface for managing your Thox.ai device. Access it via SSH or the built-in terminal in the web interface. All commands support --help for detailed usage information.
Quick Reference
| Command | Description |
|---|---|
thox start | Start the Thox.ai inference server and load models into memory. |
thox stop | Stop the running inference server and unload models from memory. |
thox status | Display current server status, loaded models, and resource usage. |
thox models | List, download, and manage AI models. |
thox run | Run a single inference query directly from the command line. |
thox config | View and modify Thox.ai configuration settings. |
thox logs | View and manage server logs. |
thox update | Check for and install firmware and software updates. |
Detailed Command Reference
thox start
Start the Thox.ai inference server and load models into memory.
Syntax
thox start [options]Options
--model, -m <name>Specify model to load on startup--port, -p <port>Port to run the server on (default: 8080)--gpu-layers <n>Number of layers to offload to GPU--context-size <n>Context window size (default: 4096)--background, -bRun server in backgroundExamples
thox startthox start -m thox-coder-7bthox start --port 3000 --context-size 8192thox stop
Stop the running inference server and unload models from memory.
Syntax
thox stop [options]Options
--force, -fForce stop without graceful shutdownExamples
thox stopthox stop --forcethox status
Display current server status, loaded models, and resource usage.
Syntax
thox status [options]Options
--jsonOutput in JSON format--verbose, -vShow detailed informationExamples
thox statusthox status --jsonthox models
List, download, and manage AI models.
Syntax
thox models <subcommand> [options]Options
listList installed modelspull <name>Download a model from the registryremove <name>Remove an installed modelinfo <name>Show model details and requirementsExamples
thox models listthox models pull thox-coder-14bthox models info thox-coder-7bthox run
Run a single inference query directly from the command line.
Syntax
thox run [options] <prompt>Options
--model, -m <name>Model to use for inference--temperature, -t <n>Sampling temperature (0.0-2.0)--max-tokens <n>Maximum tokens to generate--stream, -sStream output tokens as generatedExamples
thox run "Write a hello world in Python"thox run -m thox-coder-32b --stream "Explain this code"thox config
View and modify Thox.ai configuration settings.
Syntax
thox config <subcommand> [options]Options
get <key>Get a configuration valueset <key> <value>Set a configuration valuelistList all configuration valuesresetReset to default configurationExamples
thox config listthox config get api.portthox config set api.port 3000thox logs
View and manage server logs.
Syntax
thox logs [options]Options
--follow, -fFollow log output in real-time--lines, -n <n>Number of lines to show (default: 100)--level <level>Filter by log level (debug, info, warn, error)Examples
thox logs -fthox logs --level error -n 50thox update
Check for and install firmware and software updates.
Syntax
thox update [options]Options
--checkCheck for updates without installing--forceForce update even if up to date--channel <name>Update channel (stable, beta)Examples
thox update --checkthox updateEnvironment Variables
THOX_API_KEYAPI key for authenticated requestsTHOX_HOSTServer hostname (default: localhost)THOX_PORTServer port (default: 8080)THOX_CONFIG_DIRConfiguration directory pathRelated Articles
Need more help?
If you have questions about specific commands or use cases, our support team is here to help.