ngmi

First time #ln used, bought via boltz, received to @npub1kva…tkzv and used to buy @npub137c…k37w subscription :) now, how do i set all the relay thingies up

2026-07-17 09:37:56 CEST

Kimi K3 is now available! It is a frontier reasoning and agentic model built for coding, complex problem-solving, tool use, and long-running workflows. It supports text, image, and video inputs, structured outputs, and a 1M-token context window, making it well suited for large codebases, multimodal analysis, and tasks that need to retain extensive context.
https://nano-gpt.com/subscription/k7AXDwBb
#kimi

2026-07-17 08:04:08 CEST

- reply

Source: trust us bro

2026-07-16 18:44:36 CEST

- reply

Very cool idea. Only issue I see here is data laws. You'd have to run your own nostr nodes in the end?

2026-07-16 18:03:04 CEST

- reply

For support/form for clients we have this https://dmeta.qstorage.quilibrium.com/?nsite=nostr
👋

2026-07-16 16:18:39 CEST

The most controversial rewrite in history just shipped...
https://www.youtube.com/watch?v=CXSvKcLovAk

2026-07-16 14:26:17 CEST

- reply

Yeah I used primal as a signer. In keychain (browser inside it errors). Also I get this now in soap channel: This invite link is out of date.
The community rotated its keys after this link was made, so you're on an older version and can't read new messages. Ask whoever invited you for a fresh invite, or ask a moderator to send you the current keys.

2026-07-16 08:10:10 CEST

Does it make sense for a small company to change to #armada for comms? 🤔
#asknostr

2026-07-16 08:08:41 CEST

- reply

Keychain on android

2026-07-15 21:24:24 CEST

- reply

The soapbox channel invite errors on my end

2026-07-15 17:34:49 CEST

This is kinda cool, and comes close to #wasmbox 👏
https://nono.sh/

2026-07-15 10:30:56 CEST

- reply

i use nym on mobile, it works fine. waiting for my mullvad sub to run out to switch on desktop

2026-07-15 10:06:54 CEST

getting more and more frustrated with @npub168e…9uaw service uptimes

2026-07-14 21:39:43 CEST

- reply

Unlucky for the defense player, but yes legit

2026-07-14 08:50:25 CEST

I knew better not to touch anything with a musk stain. Gg to all #grok users, you're fucked

2026-07-13 07:59:04 CEST

- reply

Now wait til twitter takes dr too 🤣

2026-07-12 17:21:48 CEST

- reply

I'm testing https://armada.buzz/ rn

2026-07-12 15:31:18 CEST

- reply

Looks sick!

2026-07-11 18:25:48 CEST

- reply

Yes, wish it was a pwa!

2026-07-11 16:52:18 CEST

- reply

fair, but not being able to switch between the tabs there is a bug :)

2026-07-11 16:36:58 CEST

@npub175n…g6w0 for some reason notifications and dms open there too now, and i can't switch between the active ttab and articles for example. this seems wrong
https://image.nostr.build/327a57445538e6394ad20e68ebeb1164c0fb8ec97130de00b40e64af05398b06.jpg

2026-07-10 13:49:40 CEST

@npub175n…g6w0 not sure if i asked already, but can you make notes available in the right panel, so they are faster and more obvious to use? please 👋
https://image.nostr.build/a82235dedc8e86763731d9240898846aa70387c01cae8b3040c92478411978a2.jpg

2026-07-08 08:22:40 CEST

Agent Name System is totally unnecessary imo. I use npubs, there is also ENS, and both are actually decentralized. Waste of energy for the Linux Foundation
https://www.ansinfo.ai/

2026-07-07 17:13:48 CEST

- reply

Just use /advisor 👋

2026-07-06 20:45:51 CEST

Oof
#nevent1q…t4z6

2026-07-06 19:32:13 CEST

Would be kinda funny if every team in the #worldcup would withdraw now and just go home

2026-07-03 15:46:55 CEST

- reply

Something in noornote is trying to open a network ip

2026-07-03 13:43:55 CEST

- reply

noornote script block

2026-07-03 10:18:01 CEST

@npub175n…g6w0 something weird in noornote:
https://image.nostr.build/e2002bcde18db9077461fccac7182dffdf800472d2fce005b73a0710167f3dbb.jpg

2026-07-02 22:17:08 CEST

- reply

Business

2026-07-02 17:16:07 CEST

- reply

Yes, I built this https://gitworkshop.dev/[email protected]/relay.ngit.dev/nostr-web-comment-extension
Could be integrated 👋

2026-07-02 16:25:31 CEST

- reply

Switched to sidecar, much better experience than alby imo

2026-07-02 09:59:04 CEST

KYC/KYB is the worst and most annoying shit in history

2026-07-01 16:18:38 CEST

- reply

Crippled down until you pay api costs for it

2026-07-01 13:14:37 CEST

- reply

🖕🖕🖕
👇
https://z.ai/subscribe?ic=D9Z5HGTW98

2026-07-01 13:13:10 CEST

Fable is going to be redirecting coding task to Opus 4.8
🤡🤡🤡

2026-07-01 06:06:26 CEST

- reply

True, but could always podman it

2026-06-30 21:24:27 CEST

- reply

Do you have memories for chat too?

2026-06-30 21:22:19 CEST

- reply

I use claude code for cowork too 🤔

2026-06-30 18:55:26 CEST

We read a paper on agentic loops, took its one load-bearing lesson into our company of agents, and pointed the result at our own work the same day.
https://paragraph.com/@metaend/the-evaluator-is-the-floor
#airesearch #ai

2026-06-30 16:54:07 CEST

omp has a new /advisor role, and I use it currently with my free google gemini cli - confidence boost up by at least 5% IMO. pretty cool stuff

2026-06-29 20:13:09 CEST

Wait what. Captain Tsubasa is actually winning???

2026-06-29 17:52:58 CEST

- reply

Link?

2026-06-29 13:29:23 CEST

🤡🤡🤡
Time to show him the 🖕
ZAI coding CLI: https://z.ai/subscribe?ic=D9Z5HGTW98
Nano private chat + memories: https://nano-gpt.com/subscription/kUGJhgF9
Both subscriptions are still cheaper than the clowned one
#nevent1q…ymd0

2026-06-29 08:28:24 CEST

- reply

Glm subscription no, but I use nano and venice as chat already

2026-06-28 18:18:03 CEST

Is @npub1hru…f3yh post quantum ready? Or just nip17?

2026-06-28 15:11:22 CEST

- reply

It was good 2025, but nowadays it's empty. Protocol is not good either. I prefer nostr

2026-06-28 12:18:42 CEST

- reply

Works

2026-06-28 09:03:08 CEST

- reply

Oof installing an app. Any webapp?

2026-06-28 08:56:43 CEST

- reply

Link 🤔

2026-06-28 08:23:43 CEST

My plan to move away from claude chat/cli
Coding: GLM quarterly (avoid long lock in) Pro https://z.ai/subscribe?ic=D9Z5HGTW98
Chat: Venice, Private and memory https://venice.ai/chat?ref=ksiwS7 or nano https://nano-gpt.com/subscription/kUGJhgF9
Ran the numbers, and it's not economical to use a full pay per use API for coding. 😔
#ai #glm #venice
https://blossom.primal.net/5134ece2761c422eb30df8ec3ae3954306e8d1640de40359c2c5b1a04df2dd3e.jpg

2026-06-27 18:13:43 CEST

- reply

Neither in noor nor in amethyst anything

2026-06-27 17:01:53 CEST

One time dm login doesn't seem to work @npub1nxy…avr7

2026-06-27 16:49:41 CEST

- reply

No, asked my agents to implement it (custom)

2026-06-27 09:30:15 CEST

In#anthropic continues with their bs, I will probably go with the glm pro plan at $50/mth.
🚀 You’ve been invited to join the GLM Coding Plan! Enjoy full support for Claude Code, Cline, and 20+ top coding tools — starting at just $18/month. Subscribe now and grab the limited-time deal!
👉Join now: https://z.ai/subscribe?ic=D9Z5HGTW98

2026-06-27 08:44:12 CEST

Alignment: chaotic neutral
#nevent1q…qxau

2026-06-27 08:35:27 CEST

- reply

Haven't heard from them in quite some time. Usually there are updates on yt(?). Maybe they ran out of juice

2026-06-26 18:51:04 CEST

I wonder why so many apps are locked behind zapstore? Most apps could be a PWA with absolutely no lock in
#nevent1q…fygy

2026-06-26 14:22:13 CEST

- reply

I have this open, but needs more research https://shakespeare.diy/project/nym-privacy-switch

2026-06-26 13:24:06 CEST

signing fresh omp sessions now with nip7 to add a kind of human-in-the-loop proof
https://image.nostr.build/babbfb899da4a7b9633ae5f96d9cbc4090a1cb636f84a9db87356408fa9e70ee.jpg
#nostr #claude #omp

2026-06-26 07:53:52 CEST

How we run our whole company as an org chart of agents on omp, backed by a single Turso daemon, Nostr-signed decisions, and a vault that travels in the repo behind one passphrase.
https://paragraph.com/@metaend/running-a-company-as-an-org-chart-of-ai-agents
#ai #agents #nostr #turso #rust

2026-06-26 07:25:52 CEST

#gm https://blossom.primal.net/49b8e498ab52d50fd332bdea77dfa18b4f99fbe8be2602ee5a78e1307dc83eb2.jpg

2026-06-25 18:39:27 CEST

GLM 5.2 and GLM 5.2 Thinking are now cheaper to use.
https://nano-gpt.com/subscription/kUGJhgF9

2026-06-25 18:24:14 CEST

I think its kinda cool how easy it is to use nostr ids outside of nostr. Nfc tag, proof of id to and for our company agents. Wish nostr would be bigger tho
#nostr

2026-06-25 15:55:33 CEST

building a rust based TUI now to spawn and manage omp and hermes sessions. the kicker: I (or other humans) sign in via nostr, so the orchestrator knows which human its working with in the session.
#nostr #hermes #omp

2026-06-25 12:13:11 CEST

Keeping a human in the loop over a company of AI agents
What we added after going autonomous: a private assistant we reach over Nostr, and a live cockpit that shows the agents at work.
https://paragraph.com/@metaend/human-in-the-loop-over-a-company-of-agents
#nostr #agents #hernes

2026-06-25 10:46:02 CEST

My whole company runs now mostly on agents (each with their own npub) plus I reach it via HQ ui or a Hermes managed nip17 dm bridge.
#nostr #hermes

2026-06-25 08:04:26 CEST

Gm 👋 https://blossom.primal.net/1bc1a7487d3fe8914e78ad77877986d43a0e2149db76086da304a70f8a63f35c.jpg

2026-06-24 21:41:31 CEST

Sounds pretty cool?
Sakana Fugu
Frontier-level performance without single-vendor dependency. Fugu dynamically orchestrates the world's best models to tackle complex, multi-step tasks. Plug collective intelligence directly into your workflows today with a single API.
https://sakana.ai/fugu/

2026-06-24 13:50:21 CEST

NTAG424 DNA but with nsec
Drops mic

2026-06-24 12:25:59 CEST

@npub175n…g6w0 would be cool if noor could show these kinds too
https://image.nostr.build/cced36cf14dae41e001cc0b1750e7ce7e7ac862b595602907fe5c429f96eb7a4.jpg
I built a full chrome extension around it https://gitworkshop.dev/[email protected]/relay.ngit.dev/nostr-web-comment-extension

2026-06-24 10:34:39 CEST

- reply

no u

2026-06-24 10:08:11 CEST

Gm 👋 https://blossom-01.uid.ovh/38e4d22c7c1a193c05523cbc6abfff978bbb120369e6218137e370fb65c45dd0.jpeg

2026-06-24 08:16:15 CEST

Running frontier AI models locally is about to get real.
AMD's Ryzen AI Max+ PRO 495 is landing Q3 2026 with 192GB unified memory per node. Two of these give you 384GB shared CPU/GPU memory — enough to run GLM 5.2 (744B params, 40B active MoE) at 1-2 bit quantization entirely on your desk.
No cloud GPUs. No API credits. No uploading sensitive docs to OpenAI.
The setup:
- 2× PRO 495 nodes (~$9-10K total)
- 10GbE networking between nodes
- llamacpp with RPC for multi-node inference
- 384GB total memory pool
What fits:
- GLM 5.2 at 1-bit (223GB) ✓
- GLM 5.2 at 2-bit (245GB) ✓
- Future 1T+ parameter models with aggressive quant ✓
Why it's future-proof:
Unified memory architecture scales with model size. As open models keep growing, you're not buying new GPUs with fixed VRAM — you're allocating from a massive shared pool. 2 nodes today, add a 3rd later if needed.
The economics:
One-time capex vs $20-50K/year in cloud GPU rental. Pays for itself in 6-12 months for heavy users. And you own the hardware.
Cloud AI isn't going away. But for private agents, sensitive data, and uncensored local inference — this is the inflection point.
#LocalAI #AMD #StrixHalo #GLM5 #OpenSourceAI #AIInfrastructure #DataSovereignty

2026-06-23 22:51:37 CEST

is #codeberg down?

2026-06-23 14:05:52 CEST

man I hate that #codeberg randomly blocks #mullvad nodes.

2026-06-23 12:38:37 CEST

testing #notedeck - was a little bit of work necessary to make it run smooth on kde plasma - ubuntu. didn't pick up the icons etc, claude fixed it

2026-06-23 12:22:00 CEST

@npub1xts…kk5s would be cool if notedeck was available on flatpak 🥺

2026-06-23 12:04:23 CEST

- reply

Yes, I moved to omp.sh which essentially does it via orchestrate

2026-06-22 12:18:15 CEST

- reply

ah nice. will follow and get it once implemented

2026-06-22 12:16:48 CEST

Scored F (15%) for AI agent-readiness on metaend Grade. Scan your site → https://metaend-grade.fly.dev/r?host=noornote.app&grade=F&score=15
@npub175n…g6w0 👋

2026-06-22 12:15:45 CEST

- reply

i need to check out your app. it's paid only atm, isn't it?

2026-06-22 12:13:09 CEST

- reply

moved to codeberg (for private stuff) and gitworkshop for everything else arleady :)

2026-06-22 08:30:39 CEST

DO👏NOT👏KYC👏FOR👏FABLE👏
Either use multi providers like
https://nano-gpt.com/subscription/kUGJhgF9
https://ppq.ai/invite/9bc57632
https://venice.ai/chat?ref=ksiwS7
or get GLM subscription
#nevent1q…r2j9

2026-06-22 07:21:37 CEST

Fuck em not gonna happen
#nevent1q…8ac0

2026-06-21 16:56:34 CEST

- reply

Ahh. Do you know who's the dev here?

2026-06-21 16:38:48 CEST

- reply

But keychat is your app too? Sorry for the confusion

2026-06-21 15:40:01 CEST

- reply

But keychat has its own separate encryption model, doesn't it?

2026-06-21 15:10:38 CEST

- reply

"Post-quantum secure" means the encryption still holds once large quantum computers exist. The catch: a quantum computer running Shor's algorithm breaks the classical key exchange almost every secure messenger relies on, including the X25519 in Signal's X3DH handshake that Keychat is built on.
It matters now, not later, because of "harvest now, decrypt later." An adversary records your encrypted Nostr traffic today and sits on it until quantum computers can crack it. Anything private sent through Keychat this year could be opened years from now. For a messenger whose whole pitch is confidentiality, that is the real exposure.
So the actual question is: does Keychat's Signal handshake use classical X3DH (X25519 only), or the post-quantum hybrid PQXDH (X25519 plus ML-KEM/Kyber) that Signal shipped in 2023? If it is stock Double Ratchet with classical X3DH, then no, not PQ-secure yet, same as most apps. Same question for your MLS group ciphersuite, and note the Nostr secp256k1 identity keys are classical too.
The fix is the hybrid path: keep X25519, add ML-KEM, so you stay secure even if one side breaks. That is what PQXDH and the post-quantum MLS ciphersuites are for.

2026-06-21 13:47:58 CEST

Can you run a #relay in wasm/browser? That would be kinda dope

2026-06-21 12:13:16 CEST

- reply

Sounds pretty cool?

2026-06-21 11:25:42 CEST

- reply

cc @npub175n…g6w0

2026-06-21 09:19:19 CEST

Is #keychat post quantum secure? 🤔

2026-06-20 20:37:21 CEST

Testing #keychat and #noornote on mobile. This is actually working great. Was looking for an app like this!

2026-06-20 15:02:50 CEST

Cutting edge privacy stuff you can use now for cheap:
VPN: nymvpn blockchain based, 1 month free via https://nym.com/pricing?ref=31n6wmEUq62
AI: TEE with local encryption via PPQ https://ppq.ai/invite/9bc57632
Analytics: Cookieless, differentially-private, self-hosted analytics for Astro. https://gitworkshop.dev/[email protected]/relay.ngit.dev/astro-private-count
#privacy #ai #vpn

2026-06-20 14:10:38 CEST

NVIDIA AI Introduce SpatialClaw: A Training-Free Agent That Treats Code as the Action Interface for Spatial Reasoning
https://www.marktechpost.com/2026/06/19/nvidia-ai-introduce-spatialclaw-a-training-free-agent-that-treats-code-as-the-action-interface-for-spatial-reasoning/

2026-06-20 13:09:10 CEST

- reply

Depends on the tools, mine us kinda heavy too 🤔

2026-06-20 12:53:24 CEST

- reply

For omp you need around 47k ctx to be somewhat productive. 30k go for tools and system prompts