Skip to main content

A powerful terminal chat interface for multiple AI models with local session storage

Project description

TRMX - Terminal Chat Interface

A powerful terminal-based chat interface that lets you interact with various AI models directly from your command line. TRMX stores your conversations locally and makes it easy to manage multiple chat sessions.

Features

  • Seamless integration with multiple AI providers (OpenAI, Anthropic, Together, Groq, Fireworks, Cerebras, Google)
  • Easy local storage and management of chat sessions
  • Intuitive command-line interface
  • Advanced session management with titles and session IDs
  • Display of model and provider information for each chat
  • Custom time format display options
  • Auto-update capability

Installation

pip install trmx

Quick Start

# Start a new chat using your default provider and model
trmx

# List your saved chat sessions
trmx --list

# Update TRMX to the latest version
trmx --update

Detailed Usage Guide

Managing Chat Sessions

# Start a new chat session
trmx

# List all previous chat sessions
# Shows title, ID, creation date, message count, provider, and model
trmx --list

# Continue a previous session (multiple ways)
trmx 92f31c          # Using a partial session ID directly
trmx 92              # Even just a few characters will work
trmx --continue 92   # Using the --continue or -c flag
trmx -c 92           # Short form

# Delete a chat session (by its number in the list)
trmx --delete 3      # Deletes the 3rd session in the list

# Show information about chat storage location
trmx --info

Configuring Models and Providers

TRMX supports multiple AI providers including OpenAI, Anthropic, Together, Groq, Fireworks, Cerebras, and Google.

# List all available providers and their status
trmx --list-providers

# List available models for the current provider
trmx --list-models

# List models for a specific provider
trmx --list-models --provider openai   # GPT models
trmx --list-models --provider anthropic # Claude models
trmx --list-models --provider groq     # Llama and other models

TRMX now dynamically fetches the latest available models from each provider using AirTrain's ListModelsSkillFactory, ensuring you always have access to the most up-to-date model options.

# Use a specific provider and model for a single chat session
trmx --provider openai --model gpt-4
trmx --provider anthropic --model claude-3-opus-20240229
trmx --provider groq --model llama-3-70b-8192

# Set a new default provider/model configuration
trmx --add --provider openai --model gpt-4-turbo
trmx --add --provider anthropic --model claude-3-haiku-20240307

Display Settings

# Set the time display style for chat sessions
trmx --set-timestyle iso      # Display times in ISO format (2025-03-17T22:55:28)
trmx --set-timestyle human    # Display times in human-readable format (2025-03-17 22:55:28)
trmx --set-timestyle relative # Display times in relative format (2 hours ago)

# Show model's thinking process (for supported models like DeepSeek)
trmx --provider fireworks --model fireworks/deepseek-r1 --show-thinking

Maintenance

# Check the current version
trmx --version

# Update to the latest version
trmx --update

# Show help information
trmx --help

Chat Interface Features

During a chat session:

  • The provider and model information are displayed prominently
  • Chat history is shown when continuing a session
  • Type exit, quit, or q to end the session
  • Multi-line input is supported:
    • Use /m, /multiline, /multi, /p, or /paste (end with /end)
    • Use triple quotes """ or ''' (end with corresponding triple quotes)

Configuration

TRMX can be configured using environment variables:

  • TRMX_DIR: Path to store chat history, credentials, and configuration (default: ~/.trmx)
  • API key variables for each provider (e.g., OPENAI_API_KEY, ANTHROPIC_API_KEY, etc.)

You can set these in your shell or create a .env file in your working directory.

API Keys

TRMX will search for API keys in this order:

  1. Environment variables
  2. Credential files in ~/.trmx/credentials/
  3. Interactive prompt (if not found, TRMX will ask if you want to enter and save the key)

Example Configuration

For OpenAI:

export OPENAI_API_KEY=your-key-here

For Anthropic:

export ANTHROPIC_API_KEY=your-key-here

Session Information

When you list your sessions with trmx --list, you'll see:

  • Session title (auto-generated from the conversation)
  • Session ID (unique identifier)
  • Creation time
  • Message count
  • Provider (which AI service was used)
  • Model (which specific model was used)
  • Preview of the conversation

Requirements

  • Python 3.8 or higher
  • Internet connection for AI model access

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

trmx-0.3.10.tar.gz (18.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

trmx-0.3.10-py3-none-any.whl (20.1 kB view details)

Uploaded Python 3

File details

Details for the file trmx-0.3.10.tar.gz.

File metadata

  • Download URL: trmx-0.3.10.tar.gz
  • Upload date:
  • Size: 18.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.10.14

File hashes

Hashes for trmx-0.3.10.tar.gz
Algorithm Hash digest
SHA256 7c66f4ebdc05166a687e8de76e3115a1b6629be299a96dec4179273cff57d890
MD5 af94145f502da6cc8eeac2a3e2a9103a
BLAKE2b-256 a4149e5a7e1f12054476123491fdf7996d82355719915fe8c9c37b90bbee5d12

See more details on using hashes here.

File details

Details for the file trmx-0.3.10-py3-none-any.whl.

File metadata

  • Download URL: trmx-0.3.10-py3-none-any.whl
  • Upload date:
  • Size: 20.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.10.14

File hashes

Hashes for trmx-0.3.10-py3-none-any.whl
Algorithm Hash digest
SHA256 587cecd90378685cd78849f4fbee1c0448c919e055858bbf28c4b84bd1207389
MD5 78422e9d1deb4706f78c4573e1518388
BLAKE2b-256 ce59d6fef6b7e815945c22575bb767628a297fdd742ff386722d6067f3675241

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page