a billion parameters in your pocket

Joined September 2024
Over the past couple days, been making my own upgrades to @fullmoonapp since it's open source - voice mode - install models via hugging face url - ambient sounds to help maintain focus in the app - code block rendering Cool app. Dm if you want testflight
2
1
2
543
Local AI models are the future and widely underestimated. Examples of what’s already here: - Mobile: @fullmoonapp & @allen_ai’s app by @soldni run models locally on your phone - Model Training: @flwrlabs lets you train models on your local data - WebGPU means you can use @huggingface to run @deepseek_ai in your browser (see @xenovacom’s demo)
2
11
2,393
I love seeing well *designed* products being open-sourced: last week @mainframe open-sourced the fullmoon iOS app, it now supports both llama3 & deepseek
1
6
124
17,172
Using @fullmoonapp with DeepSeek locally on my Mac, makes things much more efficient and better ! This is a recommended learning path for React/NextJS within a month
1
1
4
2,029
Heard of Deepseek and want to try it out? But afraid of giving your data to China? Made a quick vid on how you can run Deepseek locally on your computer so you can keepo all your data using @fullmoonapp.
7
4
43
13,532
fullmoon v1.2.0 is now available on the App Store so you can reason free, locally, and privately on all of your devices - adds DeepSeek-R1-Distill-Qwen-1.5B - completely free and open source - works on iPhone, iPad, Mac, and Vision
17
10
3
167
36,477
fullmoon retweeted
Running DeepSeek locally on my Mac with Full Moon🌕 by @mainframe it's both private and open source. yeah it's still early and rough around the edges, but this feels like the future.
11
3
1
123
25,264
fullmoon retweeted
I always appreciate the fine details @jsngr adds to everything he builds This is fullmoon, one of the latest apps from @mainframe — it provides on-device LLMs in a simple, effective UI Notice the little fullmoon 🌕 icon used for the text streaming cursor as the LLM responds 🤌
1
1
14
3,235
deepseek-r1-distill-qwen-1.5b-4bit now available on testflight
we just added an early preview of DeepSeek-R1 to the @fullmoonapp TestFlight you can use it on iPhone, iPad, Mac, and Vision
1
22
4,063
we just added an early preview of DeepSeek-R1 to the @fullmoonapp TestFlight you can use it on iPhone, iPad, Mac, and Vision
25
15
5
415
52,947
fullmoon retweeted
Replying to @jsngr
i can feel the agi
1
1
5
1,125
fullmoon retweeted
1
1
13
3,372
fullmoon retweeted
chatting with deepseek R1 distilled 7B qwen model with @fullmoonapp
1
1
11
1,509
installing DeepSeek-R1 @fullmoonapp
🚀 DeepSeek-R1 is here! ⚡ Performance on par with OpenAI-o1 📖 Fully open-source model & technical report 🏆 MIT licensed: Distill & commercialize freely! 🌐 Website & API are live now! Try DeepThink at chat.deepseek.com today! 🐋 1/n
2
1
63
13,795
fullmoon retweeted
lil improvements to @fullmoonapp mobile web experience drawer component instead of a sidebar
4
1
11
1,699
iOS app for chatting with local LLMs
7
42
2
664
68,635
fullmoon retweeted
Use @fullmoonapp to easily have a local LLL on iPhone or mac: Unlock Local AI: FullMoon APP Brings LLMs to Your Mac & iPhone youtu.be/eB3unfQxLVM
1
3
818
fullmoon retweeted
yesterday we launched @fullmoonapp on the app store for iPhone, iPad, Mac and Vision Pro. today we are releasing a lightweight client for the web again, completely free and open source
12
14
4
223
88,534
i put @fullmoonapp on my action button for quick llama queries, no internet required. pretty sweet already more useful than “apple intelligence”
1
1
9
1,697