Pocket AI Lab
Pocket AI Lab is #140 in Developer Tools Free in South Korea and charting in 1 of 30 countries we track.
Updated 25 Oct 2026 · rankings refresh through the day
DeveloperHAOTONG TECHNOLOGY (HK) CO., LIMITED
CategoryDeveloper Tools
PriceFree
Released
Rating★ 0.0 (0)
Version1.1.0
Age rating4+
First seen by us
25 Sep 2026
25 Sep 2026Rating
0.0 ★
0 ratings
Measured
Best rank now
#140
Developer Tools FreeMeasured
Pocket AI Lab on the App Store
App details
App ID
6802253142
Publisher
HAOTONG TECHNOLOGY (HK) CO., LIMITED
On the store since
Aug 2026
Under a year
Minimum OS
iOS 26.1+
iPhone 11 and later
Devices
iPhone · iPad
Universal purchase
Content rating
4+
Languages
1
EN
Size
41 MB
Where Pocket AI Lab's downloads come from Modelled
Last 30 days: organic search, browsing the store, features and paid ads.
Where it ranks now
Every top chart it appears in across 30 countries.
| Country | Charts | Rank |
|---|---|---|
| South Korea | Developer ToolsFree | #140 |
Apps near Pocket AI Lab on the chart
Developer Tools Free, around #140.
Ratings
App Store, worldwide.
Rating
0.0 ★
0 ratings
Measured
Ratings
0
Measured
About Pocket AI Lab
Pocket AI Lab scans your device, tells you honestly which models will actually fit, downloads them straight from Hugging Face, and lets you chat with them fully offline.
Multi-backend inference - MLX, llama.cpp (GGUF), Core ML, and Apple Intelligence, switchable at runtime
Multimodal - text, images and video frames (vision models), speech-to-text (Whisper)
Import any model by link - paste a Hugging Face URL, pick a quantization, get a fit verdict before a single byte downloads
Honest memory budgets - recommendations are computed from the real per-process memory limits of your device, not marketing RAM numbers
Live metrics CPU, RAM and thermal pressure while the model is thinking
Streaming chat token-by-token, with full conversation history on every backend
Multi-backend inference - MLX, llama.cpp (GGUF), Core ML, and Apple Intelligence, switchable at runtime
Multimodal - text, images and video frames (vision models), speech-to-text (Whisper)
Import any model by link - paste a Hugging Face URL, pick a quantization, get a fit verdict before a single byte downloads
Honest memory budgets - recommendations are computed from the real per-process memory limits of your device, not marketing RAM numbers
Live metrics CPU, RAM and thermal pressure while the model is thinking
Streaming chat token-by-token, with full conversation history on every backend
Latest updates
1.1.0
New models: Gemma 4 E2B, Qwen 3.5 and LFM2 (GGUF). Long conversations keep their context in GGUF chats. Each model’s own chat template, with an inspector and a raw view of what the model receives. Benchmark suite grew to 118 tasks, with import of your own suites. Standard benchmark runs stop after 20 minutes.
1.0.1
GGUF models work again, including vision. Core ML text models load and answer.
• Long prompts no longer crash GGUF chats
• Import your own GGUF files from the Files app
• Memory budgets tuned for 6 GB and 8 GB iPhones
• New Model Benchmark with a shareable report card
Store changes
What Pocket AI Lab changed on its store page: name, icon, screenshots, price, size, languages, requirements.
Description updated 
Description updated 