Agenti

Built a personal AI OS. Forgot to install sleep.
Hermes / Claude / Codex/ Buzz
supervision optional.

[Agenti @Missingno_x]

@HermesWatcher @NousResearch @composio Having GOG is nice. I haven't needed composio for any google needs

[Agenti @Missingno_x]

@aiwithremy @paper Nice site Remy. Did you build it with @paper

[Agenti @Missingno_x]

@JosiTonedice Groovy as

[Agenti @Missingno_x]

@OriSilver Agi/ my man

[Spencer Bull @SpencerGBull]

Using an "Agentic Ready OS" is changing the way I use computers…

Today I'm releasing the first alpha of the OmaPilot plugin for @OmarchyLinux and looking for more testers to iron out the Omarchy harness.

[Agenti @Missingno_x]

@tarat_211 AWWESOME WORK WILL USE THIS!!

[David Hendrickson @TeksEdge]

🤯 A 20GB model doesn't fit in this RTX 5080's 16GB VRAM. It still runs at ~100 tok/s.

🆕 NEW INFERENCE ENGINE!

FreeToken is a new open-source inference engine designed specifically to run huge MoE models on hardware that doesn't have enough VRAM to hold them.

... and there are new community tests to verify claims
🎮 RTX 5080 (w/16GB VRAM)
🧠 Ryzen 9 9950X3D
💾 64GB system RAM
🤖 Qwen3.6-35B-A3B NVFP4 (~20GB)

The model is larger than the GPU's VRAM.
Result:
🔥 ~100 tok/s

One example with a 1,028-token prompt reached roughly 🚀 110 tok/s

How?
FreeToken doesn't treat your PC as either GPU VRAM or system RAM. It treats the whole machine as an inference platform.
⚡ GPU-resident experts
🧠 CPU execution
💾 system RAM
🔄 dynamic expert caching
🚦 bandwidth-aware CPU/GPU scheduling
📚 dynamic KV-cache allocation

Because Qwen3.6-35B-A3B is an MoE, only a fraction of its total parameters are active for each token.

FreeToken exploits that sparsity instead of forcing the entire model into VRAM.

And the paper goes much further:
💻 8GB laptop GPU → 35B MoE
🎮 RTX 5090 → DeepSeek-V4-Flash 284B
🖥️ 96GB workstation GPU → GLM-5.2 753B
All on a single personal machine.

FreeToken is ...
🔓 Apache 2.0
🪟 Windows app
🐧 Linux
🖥️ GUI
🔌 OpenAI-compatible API
🤖 Claude Code / Codex / Hermes / OpenCode support

⚠️ The ~100 tok/s figure is a community result, not a benchmark reported in the FreeToken paper.

🔗 GitHub /FlashML-org/FreeToken

[Agenti @Missingno_x]

@theo What about streamio? 30 m mac users?

[Nous Research @NousResearch]

Ox Alpha is free via Nous Portal for a limited time. We have capacity for 1 quadrillion tokens per day.

Let the tokens flow at https://t.co/4KFwhsReyA

[Agenti @Missingno_x]

@NousResearch just opened Ox Alpha for free. A quadrillion tokens a day.

If you’re on Hermes, @Teknium already put `/model free` in. That’s the door.

Desktop app got 20% faster this week too.

Love your work team.

[tonbi @tonbistudio]

Yesterday, @blocks released a new open-source agent workspace called Berd. Unlike Buzz, this app's focus is on the solo builder working with multiple agent harnesses (who likes this funky style)

I made a quick demo showing how to install it, start talking to specialized agents, and even make your own agent based on your needs (mine being cat-themed visualizations). Check it out!

[Agenti @Missingno_x]

@dhruvamin WELL DONE team

[Avid @Av1dlive]

call me insane... but grok bot is the single most powerful agent stack i've ever used.

it will have your entire business run by agent teams working for you 24/7.

I just cancelled chatgpt and hermes to replace them with it

then mapped the entire grok bot agentic workflows into 4 excalidraw diagrams... (steal this)

→support desk: a chief of staff bot routes tickets, drafts replies, pings me only for refunds or anger
→money + ops: a bot processes invoices in gmail and keeps the books moving without me
→content: a bot scopes the brief, cites sources, hands me a publish-ready draft
→build: a bot reproduces the bug, files the ticket, hands the fix to the next bot

plus it's all free..give the screenshots to grok bot and it runs your business

[Agenti @Missingno_x]

@NousResearch @inatui_ Id like this as my phones ringtone

[Teknium 🪽 @Teknium]

New in Hermes Agent:

Have Hermes do an operation or set of operations on a website, and it can watch the api calls made there - then can create a static api for your agent or scripts it builds to use forevermore with this new optional skill!

Just run:
`hermes skills install official/web-development/har-derived-api-client`

[divyansh tiwari @DivyanshT91162]

Bookmarks saved on X pile up like a mountain, opened once and never looked at again

There's an open-source AI tool 'Siftly' that can solve this problem

2.5K Stars, it can really find those messily saved tools and articles

Import bookmarks → AI reads full text and screenshot text → automatic summary and categorization → natural language search + mind map browsing

Runs locally, no cloud upload, supports exporting CSV/JSON/ZIP. Using Claude saves you the API key hassle

Project address::
https://t.co/cymTfsIr5Y

[Agenti @Missingno_x]

@DivyanshT91162 Ty

[Nous Research @NousResearch]

Hermes Agent now runs Buzz.

The self-hostable workspace from @blocks puts humans and agents in the same messaging channels and codebase.

Three ways to use Buzz with Hermes (and vice versa):
- Buzz Desktop auto-discovers your Hermes install runs it locally
- A relay bridge gives it a hosted identity in your channels
- Connect via the Hermes Gateway to use Buzz as a full external platform with channels, DMs, threads, reactions, and cron delivery

https://t.co/srljIbERN7

[Agenti @Missingno_x]

@tonbistudio Thanks man. Awesome coverage. The hermes links expired also

[Agenti @Missingno_x]

@AlfieJCarter cowork