Zero-Cost Local AI • No Subscriptions

Free Tools & Downloads

SpiritBox Basic 1.3.2 is the $0+ public download for Windows and macOS Apple Silicon for typed local chat through LM Studio; Voice Lab is a separate request-only source kit without a versioned, qualified customer release.

Download SpiritBox Basic for local chat. Voice Lab is a separate, request-only source kit, not a public or qualified customer download.

Choose a local tool or request-only source project

Basic 1.3.2 is the public $0+ download for Windows and macOS Apple Silicon. Voice Lab remains request-only and unqualified; YouTube Intelligence is a methodology demonstration.

Free Forever Entry

SpiritBox Basic

Typed local AI chat through LM Studio on Windows and macOS Apple Silicon. Choose a model supported by your LM Studio setup, converse with customizable Spirit Guides, and store conversation history locally.

  • Core chat can run locally after required setup and model acquisition
  • Compatible with LM Studio / GGUF engines
  • Core Spirit Guide Personas
  • Local Conversation History
  • Optional browser/system voice
  • No voice cloning or training in Basic or Pro
  • No required SpiritBox cloud account

Basic 1.3.2 downloads are available for Windows and macOS Apple Silicon. Both require LM Studio, a compatible local model, and its server on port 1234; no Docker is required. The compiled Windows build embeds Python and application libraries, so it does not require a separate Python installation. For Mac setup, follow the instructions included with the DMG. Intel Mac support is not offered.

Windows & macOS Apple Silicon • LM Studio and a local model required • No Docker required
Get Basic — $0+ USD SpiritBox Basic 1.3.2 • Windows ZIP / macOS Apple Silicon DMG
Request-only source

Voice Lab Standalone

A separate, request-only voice research source kit. A source ZIP exists, but it is not versioned or qualified as a customer package.

  • Source workflow for local voice-model training
  • Automatic Whisper Audio Transcription & Alignment
  • GPU-Accelerated Piper Fine-Tuning
  • Export Standalone .ONNX Voice Models
  • Pro plays compatible Piper voices; it does not include cloning or training
  • Ultimate exposes an optional trainer profile
  • Review the source and local workflow before use
Request-only tonight • No public Gumroad package • Not release-qualified
Request Voice Lab early access Access is manual and does not represent a supported customer release.
Methodology Demo

YouTube Intelligence Demo

A curated evidence demo illustrating how SpiritBox ingests transcripts, timestamps, and multi-speaker dialogues with source locators rather than hallucinating unstructured summaries.

  • Stage 0/1 Provenance Foundation
  • Subtitle & Timestamp Extraction
  • Disagreement & Source Tracking
  • No Registration or Signup Required
Ready for voice-powered conversation or full automation?
Choose SpiritBox Pro ($49 USD) for typed local-model conversations with opt-in Piper TTS output, or SpiritBox Ultimate ($199 USD) for the nine-service workstation. The Voice Lab trainer remains a separate request-only source project.
Get Pro ($49) Get Ultimate ($199)