# Clutcha — Offline AI chat, on-device LLM

> Offline AI chat, on-device LLM (iPhone/iPad app by 雷光 单.)

- Source: https://appshunter.io/ios/app/clutcha/id6780184415 (this page in markdown: same URL + `.md`)
- Developer: [雷光 单](https://appshunter.io/developer/6776143965)
- Category: Utilities, Productivity
- Price: Free with in-app purchases
- Age rating: 4+
- Requires: iOS 17.0 · 371 MB
- Languages: American English, Chinese (Simplified, China)
- Released: 2026-06-21
- Data updated: 2026-10-18

## What is Clutcha?

Clutcha AI— a smart assistant that runs entirely on your iPhone. Inference happens on-device, your conversations never leave the device, and there is no account and no data collection.
• On-device inference: chat with no network connection (the network is only used to download models)
• Upgraded to the latest MLX inference engine — steadier long conversations, more accurate multi-turn memory, more reliable tool calls
• A lightweight model is built in and ready to use; download 4B / 7B / 9B models for more capability (7 in total)
• 11 device tools: calculator, date & time, files, system, battery, translate, Safari, clipboard, reminders, location, contacts
• Web search (new): optional and off by default, with Tavily and Wikipedia — source links can be opened to verify
• Full LaTeX math and Markdown rendering: headings, lists, tables, quotes, code blocks. Turn rendering off to read the raw text
• Longer answers, less often cut short: higher output limits, a clear notice when a reply hits the limit, and one-tap continuation
• Settings and info commands open a panel: /help /status /model /set /prompt /language /tools /kb /game /downloader
• Adjustable session settings: context window, temperature, thinking budget, memory ceiling — remembered per model, so switching models keeps each model's own settings
• Sampling tuned to each model's official recommendations (Qwen3 / Qwen3.5 / Qwen2.5 / DeepSeek-R1)
• Knowledge base Q&A, plus Gomoku and chess with AI review
• Privacy first: no data collection, no tracking — your data stays on your device. Only if you turn on web search and run the tool is your query sent to the provider you choose
• Requirements: tested on iPhone 15 Pro Max (8GB); 12GB and up leaves more headroom for large models
• Pro unlocks: 4B / 7B / 9B models, Chess Coach, parameter tuning, custom prompts and web search

## Pricing and in-app purchases

Base price: Free.

**One-time purchases**

| Purchase | Price | Type |
| --- | --- | --- |
| 锦囊AI | $9.99 | one-time |


## Is Clutcha safe?

Not enough ratings yet to assess Clutcha.

## Is Clutcha legit?

Clutcha appears legitimate, with a few gaps. The AppsHunter Legitimacy Score for Clutcha is 65 out of 100 points across 7 verification signals from App Store and developer data.


## Version history (last 5 releases)

### 1.5 — 2026-10-04

v1.5 — a new engine and a refreshed reading experience

 ENGINE UPGRADE
 • Upgraded to the latest MLX inference engine — fixes around long-conversation caching, multi-turn memory drift and tool-call arguments
 • Models now load only when you start chatting, so nothing is held in memory at launch
 • A hard KV cache cap keeps long conversations and auto-compaction clear of the memory ceiling

 READING
 • Full LaTeX rendering — fractions, roots, matrices, cases and boxed answers now display correctly
 • Fixed formulas that failed to parse, and ones that vanished entirely when parsing failed (worst case now falls back to the raw text)
 • Markdown rendering: headings, lists, tables, quotes, bold/italic, links
 • Code blocks and inline code render separately; fenced code no longer mistaken for math
 • New "General Rendering" section to control LaTeX and Markdown rendering independently (turn Markdown off to copy raw text); charts and diagrams are enabled per prompt

 ANSWER LENGTH
 • Higher output limits: 0.6B / 1.7B from 768 → 1536, 4B models from 2048 → 3072
 • When a reply hits the limit you now get a clear notice and a one-tap "Continue"

 WEB SEARCH (NEW, OFF BY DEFAULT)
 • New Web Search tool (used via /tools)
 • Two sources: Tavily (bring your own API key, free tier available) or Wikipedia
 • Results are collapsed by default, with tappable source links; the network is used only when you turn this on and run the tool

 COMMANDS & PANELS
 • All commands now open a panel you can tap through: /help /status /model /set /prompt /language /tools /kb /game /downloader
 • No arguments opens the panel; arguments still work as a command (e.g. /set top_p 0.8)
 • Settings come in three scopes: sampling (temperature / top P) follows the prompt (remembered per model × prompt); thinking, context, output, memory profile and KV reuse are remembered per model; rendering, auto-compact and network are global
 • "Reset to default" now resets only the current model

 FIXES
 • /language "Auto" now really follows your input language (the first reply used to come back in English)
 • Fixed the first exchange not appearing after starting a new session — it needed one more message before showing
 • Fixed "Reset to default" being capped at a 4096 context budget
 • Fixed thinking mode silently raising the output limit — what you set is now what you get
 • Qwen3-0.6B and 1.7B no longer think by default, so replies come back faster (turn thinking back on with /think)
 • Removed the unstable VibeThinker-3B model (7B / 9B models cover the same ground)

### 1.4 — 2026-09-11

v1.4 —
• New 7B / 9B-class models: Qwen3.5-9B and DeepSeek-R1-7B, fully on-device
• Adaptive memory: context and output auto-tuned to your device, three profiles
• Thinking budget: cap reasoning length so the model wraps up and answers — faster replies
• LaTeX rendering for math formulas (inline + display)
• Cleaner command output with tables and lists
• Fixed a rare crash when the app goes to the background

### 1.3 — 2026-07-24

v1.3 Key Features:
 1. Game Modes - Gomoku (five-in-a-row) with AI commentary + Chess with AI Coach (1.7B specialized model)
 2. Models Download Manager - In-app model download with progress bar, speed indicator, hf-mirror/hf.co dual source
 3. Model Management - Delete downloaded models, Baidu Cloud import guide
 4. UI Polish - Download progress bar under memory bar, centered with speed

### 1.2.1 — 2026-07-10

v1.2.1：修复 Pro 状态重启后可能丢失 bug
------
v1.2 ：
• 自定义提示词持久化：添加的自定义提示词自动保存到 App 共享目录，从而支持角色扮演对话
• 建立会话快照：App 意外退出，或后台运行系统资源回收后，下次启动可自动恢复会话上下文
• system prompt自动注入当前时间和系统信息
• /clear 与 /new 分离：/clear 只清屏保留对话，/new 彻底重建会话
• /compact 上下文压缩时使用 AI 进行关键词摘要
• 输入框支持多行输入，长文字自动换行
• 优化 App 后台稳定性，优化 iOS 内存警告提示（可关闭）

### 1.2 — 2026-07-04

v1.2 更新内容：
• 自定义提示词持久化：添加的自定义提示词自动保存到 App 共享目录，从而支持角色扮演对话
• 建立会话快照：App 意外退出，或后台运行系统资源回收后，下次启动可自动恢复会话上下文
• system prompt自动注入当前时间和系统信息
• /clear 与 /new 分离：/clear 只清屏保留对话，/new 彻底重建会话
• /compact 上下文压缩时使用 AI 进行关键词摘要
• 输入框支持多行输入，长文字自动换行
• 优化 App 后台稳定性，优化 iOS 内存警告提示（可关闭）

## Apps similar to Clutcha

| App | Rating | Price | Category |
| --- | --- | --- | --- |
| [QG LocalAI](https://appshunter.io/ios/app/qg-localai/id6767470479) | — | Free | Utilities |
| [Kanon AI -Offline AI Assistant](https://appshunter.io/ios/app/kanon-ai-offline-ai-assistant/id6479363034) | — | Free | Utilities |
| [LocalAI](https://appshunter.io/ios/app/localai/id6774885847) | — | Free | Utilities |
| [Moonbox](https://appshunter.io/ios/app/moonbox/id6761363589) | — | Free | Utilities |
| [Jude - Private AI Assistant](https://appshunter.io/ios/app/jude-private-ai-assistant/id6758879100) | — | Free | Utilities |
| [脑花 - Agent](https://appshunter.io/ios/app/agent/id6762169454) | — | Free | Utilities |

## More apps by 雷光 单

- [兀雨書屋](https://appshunter.io/ios/app/id6776189318)
- [闲情偶寄 · 篆刻笺注](https://appshunter.io/ios/app/id6780138091)
- [vPlayer HiFi](https://appshunter.io/ios/app/vplayer-hifi/id6783141736)
- [vPlayer HiFi Classic](https://appshunter.io/ios/app/vplayer-hifi-classic/id6795004292)
- [vDarkroom](https://appshunter.io/ios/app/vdarkroom/id6806514177)

All apps by 雷光 单: https://appshunter.io/developer/6776143965

---

*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 2026-10-18.*