Open-source personal cloud OS for local AI and always-on agents. Github: https://teapaw.org/t.co/xHFNB41GqU
Singapore
Joined April 2023
- Tweets495
- Following22
- Followers5.2K
- Likes250
Pinned Tweet
๐ Olares 1.12.7 is now availableโwith Router, Lares, a broader model catalog, and updates to LarePass, more language options, and Ticket.
Learn more:
olares.com/blog/olares-1-12-โฆ
๐ข ๐ช๐ต๐ฎ๐'๐ ๐ก๐ฒ๐ ๐ผ๐ป ๐ข๐น๐ฎ๐ฟ๐ฒ๐ ๐ ๐ฎ๐ฟ๐ธ๐ฒ๐? | ๐ช๐ฒ๐ฒ๐ธ๐น๐ ๐๐ถ๐ด๐ฒ๐๐
๐ฟ Put on a favorite with Plex, or create your next song with YuE2 Studio.
Explore this weekโs Market updates, with guides in the thread. ๐
๐ฌ Turn a musical idea into a song with YuE2 Studio through Lares
olares.com/docs/use-cases/yuโฆ
๐ฌ Organize and stream your own videos and music with Plex
olares.com/docs/use-cases/plโฆ
Alright, time to get busy!
Strata 0.2.11 is now out! ๐
- Supports new Unsloth Qwen3.8-Flash-Next quants
- Layer splitting is now allowed for all configurations
- Performance increase
- App & Terminal improvements
More at: github.com/Niko1221/Strata/rโฆ
This looks very promising!
Weโre releasing new Qwen3.8-Flash-Next GGUFs with 15% higher accuracy.
Unsloth Dynamic V3.1 outperforms others by >15% on Terminal Bench, DeepSWE & more benchmarks (256K context).
Run on 32GB+ RAM setups.
GGUF: huggingface.co/unsloth/Qwen3โฆ
Blog: unsloth.ai/docs/models/qwen3โฆ
How does Qwen Flash Next at Q4_K_XL run with Strata on Olares One?
Catch Lukeโs full benchmark run๐
youtube.com/watch?v=T-dygpUaโฆ
If that gets you thinking about your own projects, our new blog takes you through installing Flash-Next from Olares Market and using it with Lares.
olares.com/blog/run-qwen3-8-โฆ
So much love! Just imagine all the cool stuff you could build with it
Here is Olares one in action
I am loving the device tbh
youtu.be/hgn4A6r5A_g?si=g1zuโฆ
do watch to know the Local AI at home! LFG
Can a local AI agent set up your home server?
Michael Klements used Hermes to turn a fresh Ubuntu mini PC into a home media and personal cloud server, with all AI inference running locally on Olares One.
A similar setup had taken him around 4 hours manually. This time, Hermes handled the main setup tasks in just 44 minutes, saving about 82% of the time.
Watch the full build๐
youtube.com/watch?v=jEp3nlvgโฆ
๐ ๐ช๐ต๐ฎ๐'๐ ๐ก๐ฒ๐ ๐ผ๐ป ๐ข๐น๐ฎ๐ฟ๐ฒ๐ ๐ ๐ฎ๐ฟ๐ธ๐ฒ๐? | ๐ช๐ฒ๐ฒ๐ธ๐น๐ ๐๐ถ๐ด๐ฒ๐๐
We just added MemPalace and RustDesk Server to the lineup. You will also find version bumps across 13 existing apps, including new NAS storage support for Immich.
Explore the setup guides below. ๐งต
๐ธ Store Immich uploads directly on your NAS
olares.com/docs/use-cases/imโฆ
๐ง Share project memory across AI agents with MemPalace
olares.com/docs/use-cases/meโฆ
If your homelab already runs smoothly, why add Olares One and how does it actually fit in?
In this video, Nick integrates Olares One as a coordinator alongside his existing servers, letting them focus on backups, DNS, and VMs while Olares One offloads 24/7 local agent workloads, health monitoring, and cross-node deployments.
Catch the full video here ๐
youtube.com/watch?v=lp-qqHXEโฆ
Can a six-year-old laptop run an intensive multi-agent development stack?
In this video, Manolo uses an aging MacBook Air to tap into Olares One as his 24/7 personal AI cloud, clocking 53 TPS on a local 27B model while keeping a full-performance draw at just 175W.
Watch the full deep dive ๐
youtube.com/watch?v=N00of-7uโฆ
