mrbizarro

mrbizarro / phosphene

Public

Local generative video panel for Apple Silicon. Wraps LTX-2 MLX, joint audio+video, one-click Pinokio install.

pinokio.computerdiscover ai-video apple-silicon generative-video local-ai ltx-video
46
3
100% credibility
Found May 11, 2026 at 46 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Phosphene provides a simple desktop interface for generating short AI videos with synchronized audio locally on Apple Silicon Macs.

How It Works

1
🔍 Discover Phosphene

You find this fun tool for making videos with matching sounds right on your Mac, no internet needed.

2
📱 Get Pinokio

Download the free Pinokio app if you don't have it, it's like an app store for creative AI tools.

3
🚀 One-click install

Paste the link into Pinokio's Discover and hit Install – it sets everything up automatically.

4
📥 Grab the magic ingredients

It downloads the special files needed to create videos (about 28 GB, but only once).

5
▶️ Launch your studio

Click Start and open the colorful panel – a simple screen with options and a gallery.

6
✨ Create your first video

Pick a style like text-to-video, type a fun idea like 'a dancing cat in a forest', and hit Generate – watch it come alive with sounds!

🎉 Enjoy your videos

Your new clips with perfect audio land in the gallery, ready to watch, share, or keep creating more.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 46 to 46 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is phosphene?

Phosphene is a local generative AI frontend for creating short videos with synced audio on Apple Silicon Macs. It wraps LTX 2.3 models in MLX for text-to-video, image-to-video, first/last-frame keyframing, and clip extension, all running offline with no API keys. One-click Pinokio install handles 20-37 GB model downloads, delivering 5-10 second clips in minutes via quality tiers from quick previews to high-res upscales.

Why is it gaining traction?

It stands out with joint audio-video diffusion for frame-accurate lip-sync and sound effects, beating silent local alternatives like Wan or ComfyUI ports. Hardware-aware tiers estimate render times by your Mac's RAM, plus LoRA support, prompt enhancement via Gemma, and lossless H.264 exports. Pinokio's seamless updates and persistent queue make local generative video feel production-ready without node graphs or shims.

Who should use this?

Apple Silicon owners prototyping local generative AI video for social clips, ads, or app demos. Indie creators needing offline local generative fill for images-to-motion without cloud costs. Devs exploring local generative engine optimization on M-series chips before scaling to paid services.

Verdict

Grab it if you're on a recent Mac—early benchmarks show it's the quickest local generative video option, but 46 stars and 1.0% credibility signal it's nascent with room for broader testing. Solid docs and Pinokio ease offset the youth; watch for Linux/Intel ports.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.