# Local AI Bench — 오프라인 온디바이스 AI 벤치마크

> 오프라인 온디바이스 AI 벤치마크 (iPhone/iPad app by Minwoo Jung.)

- Source: https://appshunter.io/ios/app/local-ai-bench/id6780884116 (this page in markdown: same URL + `.md`)
- Developer: [Minwoo Jung](https://appshunter.io/developer/1870815321)
- Category: Productivity, Utilities
- Price: Free
- Age rating: 4+
- Requires: iOS 17.0 · 31 MB
- Languages: American English
- Released: 2026-06-16
- Data updated: 2026-07-23
- User reviews in markdown: https://appshunter.io/ios/app/local-ai-bench/id6780884116/reviews.md

## What is Local AI Bench?

LocalAI Bench is an on-device benchmark app that runs AI models directly on your iPhone and measures their performance and capabilities. All inference happens inside your device — your conversations and personal data never leave it.

■ On-Device Execution
- Download open AI models (Llama, Qwen, Gemma, Phi, and more) and run them right on your device
- Works fully offline once a model is downloaded

■ Real-Time Performance Metrics
- Generation speed (tok/s), time-to-first-token (TTFT), peak memory (RAM), and thermal state — displayed live while you chat
- Automated benchmark with a standard methodology: repeated measurements of text generation, prompt processing, translation, coding, and math speed

■ Capability Evaluation
- Auto-graded tasks: math, coding, multilingual translation (Korean↔English/Japanese/Chinese), summarization, reasoning, and vision
- Compare capability, speed, and memory at a glance (choose 2-axis, composite, or efficiency scoring)

■ Vision & Speech Models
- Vision: attach photos or recognize objects through the live camera
- Speech: compare Whisper (tiny/base/small) transcription speed (RTF)

■ Compare & Manage Models
- A/B head-to-head between two models, stress tests (thermal throttling curve)
- Search models by capability (vision, coding, reasoning) or vendor and add them, with automatic iOS compatibility checks

■ Web Search (optional)
- When enabled, fetches Wikipedia references for the model — ask about recent events
- Inference stays 100% on-device; automatically skipped when offline

Pure on-device AI — no cloud APIs, no fees, no subscriptions. Runnable models depend on your device's performance; larger models are recommended for devices with more memory.


## Version history (last 3 releases)

### 1.1.0 — 2026-07-02

Release Notes
• Version 1.1.0 

Gemma 4 Support
Download and run Google's latest on-device model, Gemma 4 (E2B), and measure its performance right on your iPhone.

Web Search
Turn on the globe toggle and the app fetches relevant Wikipedia references before answering — so you can ask about events beyond the model's training data.
- AI inference still runs 100% on-device
- Automatically skipped when offline or in Airplane Mode

English Support
Switch the app language instantly from the top bar.

Improvements
- New catalog models appear immediately after app updates
- Better search accuracy for follow-up questions that continue a topic

### 1.0.1 — 2026-06-17

1.0.1 릴리즈 노트
 • 오프라인(비행기 모드)에서도 받아둔 모델로 채팅이 정상 동작하도록 수정했습니다.
 • 일부 모델에서 답변에 불필요한 번호 형식이 섞여 나오던 문제를 해결했습니다.
 • Gemma 계열 모델에서 채팅 시 발생하던 오류를 수정했습니다.

### 1.0 — 2026-06-16

No release notes.

## More apps by Minwoo Jung

- [STERO Lite](https://appshunter.io/ios/app/stero-lite/id6758091280)

All apps by Minwoo Jung: https://appshunter.io/developer/1870815321

---

*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 2026-07-23.*