SaaS Metrics
Gemini Executive Synthesis
Hitoku Draft, an open-source, voice-first, context-aware AI assistant for macOS. It runs entirely locally, supporting text generation (Gemma 4, Qwen 3.5) and multiple STT backends (Parakeet, Whisper, Qwen3-ASR).
Technical Positioning
A privacy-focused, local-first AI assistant for macOS that understands screen context, documents, and active applications to perform tasks like querying PDFs, replying to emails, and creating calendar events via voice.
SaaS Insight & Market Implications
The market for local-first AI solutions is expanding, driven by privacy concerns and the desire for offline functionality. Hitoku Draft directly addresses this by offering a context-aware, voice-first AI assistant for macOS that operates entirely on-device. Its ability to interpret screen content and active applications provides a significant productivity advantage, enabling seamless integration into daily workflows. Supporting multiple LLMs and STT backends offers flexibility, though performance and resource consumption challenges with Gemma 4 highlight the ongoing optimization required for local model deployment. This product targets power users and privacy-conscious professionals, positioning itself against cloud-dependent alternatives. The open-source nature could accelerate feature development and community adoption, establishing a strong foothold in the niche of secure, on-device AI assistance.
Proprietary Technical Taxonomy
Raw Developer Origin & Technical Request
Hacker News
Apr 14, 2026
Show HN: Hitoku Draft – context aware local macOS assistant
I am working on Hitoku Draft. An open-source, voice-first AI assistant that runs entirely locally. No cloud models, nothing leaves your machine. You press a hotkey, and you talk.It's context-aware; it reads your screen, documents, and active app to understand what you're working on. You can ask about PDFs, reply to emails, create calendar events, use web search, all by voice.It supports Gemma 4 and Qwen 3.5 for text generation, plus multiple STT backends (Parakeet, Whisper, Qwen3-ASR).Examples:- Gemma4 in action,
query a pdf document, reply to email, and the usual voice dictation (with optional polishing)I currently use it a lot with Claude Code, Obsidian and Apple Notes, or just read papers.Code: github.com/Saladino93/hitoku... of binary: hitoku.me/draft/ (free with code HITOKUHN2026)I am looking for feedback. My goal is to do AI research with clients interfacing, and I thought this is a nice little experiment I could do to iterate/fail quickly.P.S. (if anyone has tips about this)Current Gemma4 implementation (with small models) has some problems:- easy to hallucinate for long contexts, so had to reset it often. Tuned some parameters, but need to find a sweet spot.- Gemma4 with LiteRT is currently fast compared to the MLX implementation of Qwen3.5 (like 3x faster on my machine when dealing with images). But it has the price of memory spikes. I believe this is because LiteRT's WebGPU backend can allocate significantly more GPU memory than the model weights alone (I got 38GB of memory taken, for the E4B~4GB model!). I guess we need to wait for Google for this.- App size: because no official Swift package from Google yet, have to bundle some file (LiteRT dylibs) that adds ~98 MB to a previous MLX only version (total app goes from ~50 MB to ~150 MB)If any of this bothers you: use Qwen 3.5 instead (pure MLX), or wait for the upstream fixes from Google :)Otherwise, for the mid-term I plan to switch to a potentially slower, but safer, MLX version for Gemma4 (hopefully on the weekend).Developer Debate & Comments
No active discussions extracted for this entry yet.
Frequently Asked Questions
Market intelligence mapped to Hitoku Draft, an open-source, voice-first, context-aware AI assistant for macOS. It runs entirely locally, supporting text generation (Gemma 4, Qwen 3.5) and multiple STT backends (Parakeet, Whisper, Qwen3-ASR)..
How is Hitoku Draft, an open-source, voice-first, context-aware AI assistant for macOS. It runs entirely locally, supporting text generation (Gemma 4, Qwen 3.5) and multiple STT backends (Parakeet, Whisper, Qwen3-ASR). positioned in the market?
Based on our AI analysis of the original developer request, its primary technical positioning is: A privacy-focused, local-first AI assistant for macOS that understands screen context, documents, and active applications to perform tasks like querying PDFs, replying to emails, and creating calendar events via voice.
Which technical concepts are associated with Hitoku Draft, an open-source, voice-first, context-aware AI assistant for macOS. It runs entirely locally, supporting text generation (Gemma 4, Qwen 3.5) and multiple STT backends (Parakeet, Whisper, Qwen3-ASR).?
Our proprietary extraction maps Hitoku Draft, an open-source, voice-first, context-aware AI assistant for macOS. It runs entirely locally, supporting text generation (Gemma 4, Qwen 3.5) and multiple STT backends (Parakeet, Whisper, Qwen3-ASR). to adjacent architectural concepts including Open-source, voice-first AI assistant, runs entirely locally, no cloud models.
Are developers creating tools for Hitoku Draft, an open-source, voice-first, context-aware AI assistant for macOS. It runs entirely locally, supporting text generation (Gemma 4, Qwen 3.5) and multiple STT backends (Parakeet, Whisper, Qwen3-ASR).?
Yes, open-source adoption is correlated. An active project titled 'fikrikarim/parlor' explores similar frameworks: On-device, real-time multimodal AI. Have natural voice and vision conversations with an AI that runs entirely on your machine. Powered by Gemma 4 E...
Engagement Signals
Cross-Market Term Frequency
Quantifies the cross-market adoption of foundational terms like Claude Code and documents by tracking occurrence frequency across active SaaS architectures and enterprise developer debates.