# Models Zoo — On-device AI model zoo

> On-device AI model playground for experimentation and developer integration. (iPhone/iPad app by Daisuke Majima.)

- Source: https://appshunter.io/ios/app/models-zoo/id6762083207 (this page in markdown: same URL + `.md`)
- Developer: [Daisuke Majima](https://appshunter.io/developer/1350309854)
- Category: Developer Tools, Graphics & Design
- Price: Free
- Rating: 2.00/5 from 1 App Store ratings
- Age rating: 4+
- Requires: iOS 18.0 · 13 MB
- Languages: American English
- Released: 2026-04-14
- Data updated: 2026-08-28
- Monetization: free
- User reviews in markdown: https://appshunter.io/ios/app/models-zoo/id6762083207/reviews.md

## What is Models Zoo?

Core ML Models is a curated, on-device AI playground. Download any model from the built-in catalog and try it out instantly — no server, no account, nothing leaves your device.
                                                                                    
Every model is an open-source release converted from PyTorch to Apple's Core ML format and tuned to run efficiently on the Neural Engine, GPU, or CPU.
                                                                                    
WHAT YOU CAN DO                                                                   
• Chat with small language models (Gemma, etc.) including multimodal vision+text
• Generate images from text with Hyper-SD                                         
• Remove backgrounds, matte video subjects, and colorize old photos               
• Estimate depth and reconstruct 3D faces from a single photo                     
• Detect and classify with YOLO, SigLIP, and zero-shot open-vocabulary models     
• Separate music stems with Demucs and transcribe audio to MIDI with Basic Pitch  
• Clone voices and synthesize speech with Kokoro and OpenVoice                    
• Super-resolve images with SinSR and RealESRGAN.  

DEVELOPER TOOLS, NOT JUST A DEMO                                
• Inspect — see every .mlpackage's input/output tensor shapes, data types, compute-unit setting, on-disk size, and Core ML metadata at a glance.                           
• Benchmark — run warmup + timed iterations on your own device and get average / median / min / max latency plus memory delta. Switch compute units (All, CPU+GPU, CPU+ANE, CPU only) to find the fastest config for your hardware.                                                                                  
• Swift Code — export a ready-to-paste Core ML loading + inference snippet  wired to the model's actual input/output names, so integration into your own app is copy-paste fast.  
                                                                                    
DESIGNED FOR THE NEURAL ENGINE                                                    
Each model ships with compute-unit settings tuned per-architecture (Neural Engine,
GPU, CPU) so you get the best inference speed Apple Silicon can deliver.         
                                                                  
OPEN AND TRANSPARENT                                                              
Every model in the catalog links back to its original paper, repository, and license. Conversion scripts are public on GitHub so you can see exactly how the  .mlpackage files were produced.                                 
                                                                                    
PRIVACY FIRST                                                                     
All inference runs locally on your device. The app only fetches model files from public storage over HTTPS — there are no analytics, no ads, and no account to create.                                                         
                                                                                    
REQUIREMENTS                                                                      
• iPhone or iPad with Apple Silicon recommended for best performance
• Some large models require several GB of free storage and ≥ 6 GB RAM

## Key features

- On-device AI model catalog
- Chat with language and vision models
- Image generation and manipulation
- Audio processing and speech synthesis
- Model inspection and benchmarking
- Swift code export for integration
- Privacy-focused local inference


## Frequently asked questions about Models Zoo

### What can I do with Models Zoo?

Models Zoo allows you to chat with language models, generate images from text, remove backgrounds from videos, estimate 3D faces, detect objects, separate music stems, clone voices, and enhance image resolution, all on your device.

### Is Models Zoo free to use?

Yes, Models Zoo is completely free to download and use. There are no in-app purchases or advertisements.

### Does Models Zoo require an internet connection?

No, Models Zoo is designed for on-device inference. You only need an internet connection to download the model files from public storage.

### How does Models Zoo help developers?

Developers can inspect model tensor shapes, benchmark performance on their device, and export Swift code snippets for easy integration of AI models into their own applications.

### What kind of AI models are available in Models Zoo?

The app features a curated catalog of open-source AI models converted to Apple's Core ML format, including language models like Gemma, image generators like Hyper-SD, and object detection models like YOLO.

### Is my data safe with Models Zoo?

Yes, privacy is a top priority. All AI inference runs locally on your device, and no data leaves your device. The app does not collect analytics or require an account.

### What devices support Models Zoo?

Models Zoo is available for iPhone and iPad devices.

### How often is Models Zoo updated?

The latest version is 3.0, and it was last updated on April 25, 2026, indicating active development and regular updates.

## Version history (last 3 releases)

### 3.0 — 2026-04-25

New models have been added.
Depth Anything 3: relative monocular depth
Nitro-E: Fast image generation
Gemma 4 E4B: Larger variant
Qwen3.5 2B: LLM
Qwen3-VL 2B: Multi-modal LLM
Storage management functionality has been added.

### 2.0 — 2026-04-17

• Benchmark: measure each model's latency (avg / median / min / max) and  memory delta on your own device.                                                                    
• Model Inspector: browse every file's inputs, outputs, shapes, dtypes, metadata, and disk size. Copy a ready-to-paste Swift snippet.                                       
• 6 new models: Real-ESRGAN 4×, GFPGAN, RF-DETR Nano, Face Parsing, MobileSAM, LaMa.                                
• Real-time camera input for object detection and 3D face reconstruction.                             
• Video auto-matting: MatAnyone now pulls a first-frame subject mask from  Apple's Vision person-segmentation before refinement.                                               
• Faster model switching — models compile eagerly and are cached across  demos, with load / inference time shown on each screen. 
• Audio input for Gemma4.

### 1.0 — 2026-04-14

No release notes.

## More apps by Daisuke Majima

- [かわいいタイプ診断BeautyCaption](https://appshunter.io/ios/app/beautycaption/id1485839081)
- [aiLike](https://appshunter.io/ios/app/ailike/id1485935927)
- [ARCamera.](https://appshunter.io/ios/app/arcamera/id1488699740)
- [パイウェッブpiWeb](https://appshunter.io/ios/app/piweb/id1491581993)
- [SmilePractice](https://appshunter.io/ios/app/smilepractice/id1497464407)
- [PenguinSlipInTheDisplay](https://appshunter.io/ios/app/penguinslipinthedisplay/id1497815209)

All apps by Daisuke Majima: https://appshunter.io/developer/1350309854

## Related topics

[models zoo](https://appshunter.io/ios/topics/models-zoo)

---

*Data collected daily from the US App Store and indexed by [AppsHunter](https://appshunter.io/). User reviews are verbatim App Store reviews. Ratings, prices and chart positions refresh continuously; this snapshot is from 2026-08-28.*
