Osaurus

Open source agents that run 100% locally on your Mac

Open Source

Privacy

Artificial Intelligence

GitHub

Visit WebsiteSee on Product HuntGithub ⧉Twitter ⧉

The approval gate + local memory combo is the interesting bit for me. I’ve been nervous about agents touching real files; seeing every action before it runs feels like the right tradeoff. Curious how you handle rollback when an agent writes bad files?

persistent memory + fully offline is where the loop gets tight — the summarizer that decides what to keep is the same small model that produced the noise. compression drift compounds on itself with no bigger model to arbitrate. that's the part i'd want to see the eviction policy for, not the model swap.

Huge congrats on the launch, @tpae ! Love the offline-first approach. One thing I’d find useful: a visual activity log to scrub through what the agent did locally. Any plans for that?

Mac-native and local is the combo I keep wanting, the Firefox-of-local-AI pitch landed. Practical question: what's the real memory footprint running one of the mid-size models on, say, a 16GB machine, does it stay usable or does it eat everything and make me quit my other apps? That's usually where "runs locally" turns into "runs locally if you bought the maxed-out box." Nice to see something leaning open source instead of another closed wrapper.

Swift-native, fully offline, and open source is a rare combo — most agent harnesses assume you're fine piping everything to the cloud. The cryptographic identity piece is intriguing; I'd love to hear more about how persistent memory works across sessions and whether it's local-only storage. This feels built for people who actually care about owning their stack. Nice work.

Congrats on the launch. The approval gate feels like the right trust boundary for an agent that can touch files, Calendar, Contacts, and iMessage. Iâm curious how you think about permissions once someone has several agents or projects: are you aiming for per-action approvals only, or eventually a project-scoped policy where an agent can access one context but not another?

this is one of the more compelling "local AI agent" pitches I've seen, the approval gate before every action is the part that actually matters to me, most of these tools just yolo the agent loose on your filesystem. question on the self-scheduling piece you mentioned - does that still fire if the Mac is asleep or the lid is closed, or does it need to actively be awake/plugged in for scheduled runs to trigger?

the approval gate plus "always allow" per action is the right default, but the part I'm actually curious about is the open source contribution angle - if community members start shipping shareable agent skills/plugins that get access to iMessage or Contacts, is there any review step before those get trusted, or is it on the user to audit what a downloaded skill can touch before installing it

Local-first agents is the direction I keep hoping wins. The cryptographic identity part is interesting — most agent frameworks treat identity as an afterthought. Curious how heavy it gets memory-wise with a couple of agents running on an M-series Air?

Just discovered @Osaurus on X

• Used my ChatGPT Pro sub

• Recognised my local MLX models from LM Studio

• 💜 Dino avatars

Nice. If Qwen (or some of its models) are still open-source, which is my guess, I can confirm it works fine on a Mac Mini M3 for doing automation tasks on websites, DOM manipulation...

Downloading to check this out now! Just the branding of the dinosaurs is unique enough to pique my interest, would love to know where you got the idea to use them. Make the internet fun again!

Congrats on the launch! With the approval gate on every action, does that mean autonomous runs stop and wait a lot, or is it smart about only flagging the risky stuff like sending a message or running code?

You describe Osaurus as an AI agent platform rather than just an assistant. Long term, do you envision users interacting with a single persistent agent that accumulates years of context, or a collection of specialized agents with separate memory and identities?

The Mac-native angle is the part that stands out. Ollama works, but it always feels like it is renting space on the machine rather than living on it. What I would want to know is how model management holds up once people start pulling in the bigger local models, is there a clean way to swap them without eating all the RAM. Either way, local-first AI getting a friendlier front door is good for the whole space.

Using Osaurus every day and really appreciate all the hard work. It's my go-to for when I share Venice API keys with family and friends to introduce them to AI. I'm looking forward to being able to use image gen and TTS with a cloud or local server url. I'm using Venice for LLM in Osaurus, and it would be nice to use their image and speech models as well. I'm also setting up my 5090 GPU PC with image gen and fish audio TTS, and if it could be a local network server for Osaurus image and speech gen like it is for local LLM, that would be great for local!