That's the idea behind Knemora, an open-source, local-first AI Second Brain I've been building in my spare time.
Imagine turning your documents into interconnected knowledge, exploring relationships, and asking an AI questions about everything you've collected.
All running locally, on your own machine.
A few weeks after releasing the first alpha, I realized something:
Building an AI application is one thing. Making it genuinely useful every day is another.
So I kept building.
Knemora v0.2.0-alpha is here!
This update focuses on making Knemora more reliable, responsive, and easier to use.
Here's what's new:
- Better conversation continuity and automatic language matching.
- Wiki and Direct Chat Modes.
- Fast and Deep reasoning controls.
- Stronger knowledge compilation and relationships.
- Gemma 4 E4B joins the tested local model registry.
I also ran a small benchmark comparing Qwen3-8B and Gemma 4 E4B.
In this particular Direct-mode test, Gemma delivered 20.6% lower median application latency and 31.1% higher decode throughput.
Not a universal benchmark, just another step toward making local AI more responsive.
Knemora is still in alpha, and there's plenty left to improve. I'm building it one iteration at a time, with the goal of reaching a stable v1.0.0.
Would you use a local AI to build your own connected knowledge base? What would you expect it to do?
GitHub: https://github.com/federicolaterra/knemora
Website: https://knemora.space
Watch the demo: https://www.youtube.com/watch?v=kl5qNrWPOnY
Feedback and contributions are always welcome!
Top comments (4)
On the second question: what usually decides whether people trust a tool like this is that every answer shows which notes it came from, so you can check it, and that it never changes or deletes a note without showing the change first. Plain files on disk that still make sense if the app is gone matter a lot too. After that, the most useful everyday job is a boring one: sorting new notes into the right place and suggesting links between them, with you approving each move.
Thanks for sharing your thoughts! I completely agree, especially about transparency, data ownership, and keeping humans in control.
These are actually some of the core principles behind Knemora. Every knowledge claim maintains provenance back to its original source, and the generated Wiki consists of portable Markdown files that remain accessible independently of the application.
Knemora also preserves original documents and builds a native knowledge graph by identifying and validating relationships between entities.
I really appreciate the feedback. It's exactly the kind of discussion I was hoping to encourage by making Knemora open-source!
Good to hear, and open source is the right call for this kind of tool. One thing I'd be curious about: when a claim in the wiki is wrong, how does a person correct it, and does the correction survive the next regeneration? That is where most of these systems lose people's trust.
Knemora already implements a few safeguards around this: each claim is linked to its original source through provenance, claim identities are preserved across reprocessing where possible, conflicting evidence is retained rather than silently overwritten, and unchanged knowledge isn't unnecessarily regenerated.
However, a dedicated human correction workflow, where users can explicitly correct or approve a claim and have that decision protected across future regenerations, isn't implemented yet.
It could be a future implementation...