Apollo: Private, local AI now powered by Liquid AI
135 posts
Apollo: Private, local AI now powered by Liquid AI
@apolloaiapp
Run local models right on your phone with no internet or connect with OpenRouter. openintelligence.group
Hey everyone, if you’re running into an issue where a local model says “Loading model…” for a very long time it might not be downloaded yet. Should take 3-4 seconds at most to load. We’ll get this bug fixed soon, for now you can check to make sure it’s downloaded from Settings- LFM2 are now available on Apollo. Thanks to the @liquidai team for driving forward months of app and user improvements, we are now a part of the Liquid AI ecosystem. With that comes stronger features, faster models, and a robust roadmap. More to come soon.Today, we release the 2nd generation of our Liquid foundation models, LFM2. LFM2 set the bar for quality, speed, and memory efficiency in on-device AI. Built for edge devices like phones, laptops, AI PCs, cars, wearables, satellites, and robots, LFM2 delivers the fastest
- Apollo 1.0.1 out now. ☉ Minor bug fixes and layout improvements.
- Apollo 1.0.10 out now. ☉ Some OpenRouter errors weren’t surfacing properly. If you’ve noticed some models not working, it should now be clear why. ☉ Long-press actions. You can now long-press the app icon to open your last chat or a new chat. ☉ And other minor bug fixes.
- Apollo 1.0.2 out now. ☉ You can now specify a model identifier with your custom backends.
- Replying to @OfficialLoganKThis may not be the sort of answer you're looking for, but currently it's @apolloaiapp, simply because it so cleanly runs some basic models right on my device, with good performance, even without an internet connection. Absolutely game changing for regular people.
- Use any model from your phone.Replying to @localghostapp's OpenRouter login flow is pure magic 🤌✨
- Apollo 1.0.3 out now. ☉ Better, more obvious error handling. ☉ Models you use now sort to the top. ☉ Models now have a glyph to help you identify them.
- local ai models are now on apollo. these run entirely on your phone, preserving your privacy they also work without an internet connection, ensuring you have access to intelligence even if you’re offline or off-grid
00:00 - Apollo 1.0.14 is out and fixes a few small interface related issues: ☉ Prompts and models should now set properly. ☉ Improved behavior of some setttings menus.
- Apollo 1.0.13 out now. ☉ Local model support. Chat with Llama 3.2 1B & 3B completely privately. ☉ Redesigned chat settings. Easily set your prompt or model right from the chat. ☉ Theme & font options. Change the font, font size and more from the theme menu.
- Apollo 1.0.5 out now. ☉ Voice messages are now streamed and should be much faster to reply. ☉ Other minor fixes and optimizations.
