Own your AI. Agents that remember, execute code safely, and stay reachable from anywhere -- all on your computer. Any model. No cloud required. Open source.
- Tweets 1,409
- Following 32
- Followers 10,760
- Likes 1,291
Pinned Tweet
Models get commoditized.
The harness compounds.
Thanks @SarahPerezTC for the writeup.
Osaurus brings both local and cloud AI models to your Mac techcrunch.com/2026/05/15/os…
7
8
68
34,956
Osaurus retweeted
I'm not playing around. Since releasing the beta yesterday, I've already added:
- Persistent memory
- Agent sandboxes (using WSL Containers)
- Automations (folder watchers, scheduled tasks)
- Bug fixes
And there's lots more coming.
Osaurus Beta for Windows has support for WSL Containers on day one of its GA.
Watch a local model safely experiment with python within a sandbox to generate a Mandelbrot fractal.
Fully native, built for Windows with WinUI, C#, and .NET 10
1
4
8
1,809
Osaurus Beta for Windows has support for WSL Containers on day one of its GA.
Watch a local model safely experiment with python within a sandbox to generate a Mandelbrot fractal.
Fully native, built for Windows with WinUI, C#, and .NET 10
1
2
23
3,927
- Every agent gets its own home inside the container. One agent can't read another's files or the rest of your PC.
- The container image is built on your machine from a Dockerfile.
- API keys and other secrets never enter the conversation. They're injected into commands as environment variables and scrubbed from output.
- Network is off by default for the sandbox, or limited to an allowlist you set per agent. Traffic goes through a host proxy that checks each address.
- The agent can install packages (apk, pip, npm) and run long jobs in the background, with a Terminate button on the card.
2
2
244
Join the beta today to test it out.
osaurus.ai/windows
2
168
Osaurus retweeted
This is why we're building @OsaurusAI
You own your data.
You own your inference (choose between local/cloud)
You don't have to worry about someone training on your data, even if they claim they aren't.
With local AI, your data stays on your device.
Don't use this
Build your own software using their tokens
So when their tokens become too expensive
You can switch to your own GPUs
Your compute. That you control
So you yourself can't be controlled
You need to control your stack -- or you become enslaved
2
6
598
Osaurus retweeted
Osaurus is now on Windows 11.
Local models, private cloud access, and agents that can actually do things, built natively for your PC.
Osaurus is now on Windows 11.
Local models, private cloud access, and agents that can actually do things, built natively for your PC.
osaurus.ai/blog/windows-beta
1
4
635
Osaurus is now on Windows 11.
Local models, private cloud access, and agents that can actually do things, built natively for your PC.
osaurus.ai/blog/windows-beta
5
8
50
3,711
What's in the beta:
• Run models locally, including our Raptor 0.6
• Use Osaurus Cloud credits for private access to frontier models, or bring your own keys
• Pair with Osaurus on your Mac and share agents between them
• Workspaces, across Mac and PC
1
1
3
361
The beta is rolling out in waves. Sign up to be next: osaurus.ai/windows
Some things will break. Tell us what does: community.osaurus.ai/c/suppo…
1
3
229
Most local AI demos run on maxed-out Macs.
40% of our users have 16GB or less.
Raptor is for them. 4B. Wi-Fi off.
9
6
149
10,924
Raptor 0.6, 4B, JANG_6M quant.
Reads xlsx and docx through native file tools. Around 40 tok/s.
Built by @dealignai.
osaurus.ai/models/raptor-06
1
12
913
Osaurus retweeted
#design #ux #ai #localai
I just had a great conversation with the founder of @OsaurusAI
I’ve been following the evolution of the app since the beginning of the year. They built Raptor models, and the results are impressive.
osaurus.ai
1
1
5
380
Osaurus retweeted
Benchmarks measure what a model knows. Osaurus shows what it can do
Bonsai 2 dropped hours ago. It's already doing real work in Osaurus.
Read two CSVs. Wrote the Python. Drew the chart.
Then found the one channel losing money and told me why.
27B. Ternary. Nothing left the Mac.
1
2
416
Osaurus retweeted
A little verbose for my taste, but absolutely gets the job done. Try out the ~6gb and ~8gb quants of Bonsai 2 27b on your Mac!
Bonsai 2 dropped hours ago. It's already doing real work in Osaurus.
Read two CSVs. Wrote the Python. Drew the chart.
Then found the one channel losing money and told me why.
27B. Ternary. Nothing left the Mac.
2
3
50
6,412
Bonsai 2 dropped hours ago. It's already doing real work in Osaurus.
Read two CSVs. Wrote the Python. Drew the chart.
Then found the one channel losing money and told me why.
27B. Ternary. Nothing left the Mac.
Today, we’re announcing Ternary Bonsai 2 27B.
Based on Qwen3.8 27B, Bonsai 2 27B is 9x smaller than its full-precision counterpart while retaining 98.2% of its aggregate benchmark performance.
Two months after the first Bonsai 27B release, the biggest change is quality. The footprint remains 5.9 GB, but the gap to full precision has narrowed materially, with particularly strong gains in agentic coding, multimodal reasoning, and long-horizon tool use.
Ternary Bonsai 2 27B is available today under Apache 2.0.
10
21
336
34,657
PrismML said the big Bonsai 2 gains were agentic coding and tool use.
This is what that looks like on a Mac.
The model wrote the code, ran it in an isolated sandbox, and handed back a real PNG. Not a code block. A file.
Weights, ready for Osaurus:
huggingface.co/OsaurusAI/Bon…
2
16
1,580
JANG conversion by @dealignai. Model by @PrismML.
Osaurus is free, open source, and runs this on your Mac today.
github.com/osaurus-ai/osauru…
8
1,032
Local AI shouldn't be limited to expensive hardware. Everyone should have the opportunity to own their AI.
Raptor 0.6 just shipped.
Most Macs have 8 or 16 GB. Most AI models pretend they don't. Raptor is our agent model built for those machines. Reads folders, checks spreadsheets, drafts emails. Local, free.
3.7 GB, down 41% from the last version.
2
5
684
Osaurus retweeted
This will be coming to Osaurus for Windows as well. Not everyone has an RTX 30/40/5090. It'll open up local model usage for PCs with more common GPUs. You won't need a workstation to use local AI.
Raptor 0.6 just shipped.
Most Macs have 8 or 16 GB. Most AI models pretend they don't. Raptor is our agent model built for those machines. Reads folders, checks spreadsheets, drafts emails. Local, free.
3.7 GB, down 41% from the last version.
1
1
7
1,257