Elite Intel

⚙️ INSTALL | 🎓 WIKI | 💻TECH STUFF

---

Can be ran 100% off-line or with various cloud services

Windows 🪟

1. Download the 👉installer👈.
2. Run the installer and follow the on-screen prompts.

Linux 🐧

• Do not use zip file from the artifacts. That is for updates.
• NO sudo required!

1. Download the installer script:

curl -L -o installer.sh https://raw.githubusercontent.com/stone-alex/EliteIntel/refs/heads/master/distribution/installer.sh

2. Make the script executable and run it:

chmod +x installer.sh
./installer.sh

The app installs to ~/.var/app/elite.intel.app. Setup complete. Look for shortcut under games or utilities called Elite Intel.

👉Free cloud if you don't have hardware for local LLM👈

• You may experience some occasional timeouts or lag, but for the most part it is fine.

Paid cloud services

• Claude | claude-haiku-4-5
• Grok | grok-4-1-fast-non-reasoning
• Open AI | gpt-5.2
• Google Gemini | gemini-3.1-flash-lite
• Deepseek | deepseek-v4-flash

The cost will very depending on which one you choose.

Local LLM

• LMStudio (Local off-line Model matrixportalx/tulu-3.1-8b-supernova)
• Ollama (Local off-line Model matrixportalx/tulu-3.1-8b-supernova)

Why tulu-3.1-8b Supernova specifically

Minimum requirements for running the game and the Local LLM on the same GPU

• NVIDIA RTX 3090 24GB VRAM or equivalent.

• RTX 5090 with only 12GB VRAM will not do. Not enough VRAM.

  • Game takes 8GB (or more)
  • LLM takes 6.5GB See Configure the app for next steps.

---

Overview

✅ The off-line TTS. (Linux and Windows) Powered by Kokoro. built-in / included.

✅ The off-line STT. (Linux and Windows) Powered by NVIDIA Parakeet. built-in / included.

✅ Local off-line private LLM Installing local LLM

Elite Intel is your AI-powered, sidekick for Elite Dangerous. It's a Quality of Life (QoL) app that enhances gameplay with voice-activated commands uttered in natural language,

real-time journal parsing, and smart AI assistance-all while staying fully compliant with Frontier Developments' Terms of Service. No AFK play, and no unfair advantages; everything requires your input.

The app uses Speech-to-Text (STT) and Text-to-Speech (TTS) for natural voice interactions, and LLM for handling commands, queries, and chats. It processes game data from journal and auxiliary files, pulls in info from EDSM when needed, and keeps things responsive.

---

Contributing

Fork it and submit pull requests! Keep things modular and event-driven, and test your changes. For dev details, check out DEVELOPERS.md in the root. It covers the architecture (decoupled, multithreaded, event registry via reflection) and principles like DRY and SRP. All PRs go through review.

License

The project is open-source under a Creative Commons license CC BY-NC-SA 4.0, so feel free to check out the code and contribute. (see DEVELOPERS.md for details)

Contact Developer / Join the Community

If you run into issues, hit us up on GitHub issues or Feedback and bug reports are super welcome! To get in touch: 👉Matrix👈