N0Fears (@fears_n0) header
N0Fears (@fears_n0) profile picture

Founder @ N0Fears Labs Building local-first AI Desktop products for private use and self-hosted control

Joined February 2023
Photo by N0Fears (@fears_n0)Photo by N0Fears (@fears_n0)Photo by N0Fears (@fears_n0)Photo by N0Fears (@fears_n0)Photo by N0Fears (@fears_n0)Photo by N0Fears (@fears_n0)Photo by N0Fears (@fears_n0)Photo by N0Fears (@fears_n0)Photo by N0Fears (@fears_n0)
N0Fears retweeted
Sol is our new flagship and a step function better than GPT-5.5. Terra delivers performance competitive to GPT-5.5 at 2x lower cost. Luna is our most cost-efficient model, delivering strong capability at our lowest cost. Together, the GPT-5.6 family gives people and developers more choice in how they balance intelligence, speed, and cost.
228
372
543
5,158
2,905,519
Work should not stop inside chat history. N0 Local lets you run AI on your own computer, save useful outputs as notes, tasks, or drafts, assemble them into a document, and export the final result. Private. Local. Built for real work.
2
2
13
26,238
No tokens, no usage fees. You run it on your own machine.
77
Work should not stop just because the internet disappeared. Here is N0 Local Workspace in action. Attach files, ask AI questions, save useful output, and keep moving — all on your own machine, even when you are offline.
4
1
10
14,191
What I wanted to fix was this: chat is good for generating, but bad for keeping useful results usable. Workspace mode pushes the useful part into notes, tasks, and drafts instead of letting it die in the thread.
74
N0 Local now includes Workspace mode. Attach files, ask AI about them, and save useful output as notes, tasks, and drafts. Chat on the left. Working material on the right.
1
1,793
Choosing a local model is easy. Choosing one your computer can actually handle is the hard part. That’s what N0 Local is built to simplify. Install models, switch between them, remove what you don’t need, and keep everything fully local. More info: n0fears.com Join the waitlist to be first at release. #LocalAI #BuildInPublic
4
1
2
6
258
A lot of local AI still expects users to figure everything out themselves. I think that part should be easier.
1
46
Most users don’t need more commands. They need a product.
1
49
Everything here runs on the current machine - model selection, install flow, switching, and local chat history.
1
65
Local AI gets better when model selection stops feeling random. Working on that in N0 Local now - with hardware-aware guidance built in. Demo soon.
1
42
Your AI shouldn’t live in someone else’s tab. Built the first demo clip for N0 Local - private desktop AI on your own computer. Open it, use it, close it, come back to it. Don’t miss early access: n0fears.com #LocalAI #BuildInPublic #AIProducts
3
4
134
Still early, but the direction is clear: private, desktop-first, local-first AI.
2
28
AI on your computer should feel like something you actually own, not something you temporarily access.
2
40
Most AI agents run directly on your laptop. That’s risky. I built a setup where: • a local LLM runs on the PC • agents execute on a VPS • OpenClaw orchestrates everything Local AI + secure remote agents. Quick demo below 👇 #OpenClaw #AI #BuildInPublic #LocalAI
2
1
7
409
local LLM → SSH tunnel → VPS → OpenClaw agents
2
65
If you want to use OpenClaw securely with local AI and keep your PC isolated, this might help. This tool can: • runs a local LLM • connects it to a VPS • launches OpenClaw agents remotely No paid APIs. Local AI + secure remote agents. #OpenClaw #AI #BuildInPublic #LocalAI
2
1
1
7
508
Why this exists: Running AI agents directly on your laptop is risky. This setup: • keeps your PC isolated • runs agents on a VPS • still uses your local model So you get local AI + remote execution.
4
74