jAIsus

@maxktz Interested to know what is bad about Bitwarden ? Selfhosted it the last 5+ years and has always delivered.

switched from bitwarden to this and it's sooo better

Hermes: I need a bit more clarity to avoid guessing and breaking your setup. You said "in /skills". Could you confirm which one you mean? 1. ~/.hermes/skills/ — current profile's (default) skills folder. 2. ~/.hermes/profiles//skills/ — a specific named profile. 3. A new top-level /skills directory at the filesystem root (/skills), outside Hermes. 4. Some other path — tell me the absolute path. Also, how many profiles/agents do you have? I only see the default profile active. If there are named profiles, I need to know which ones should get the skill.

Pour ceux qui ont la flemme de retrouver le tweet, mets ça dans ton Hermes Agent : "/learn github.com/firecrawl/pdf-…" Puis: "Mets ce Skill au dossier de skills partagés et attribue-le à tous les agents de ma config qui peuvent en avoir une utilité un jour".

What an experience, Deepseek have changed the game for real. The price bar has been set for legit inference, it can only get better for both con and prosumers. Mix this with Hermes agent and you feel nothing is impossible.

@MichaelGannotti @NousResearch glad you found it Mike, its a game changer for inference everywhere.

This is such a great option and cannot believe I didnt know about it. Install @NousResearch Hermes on any laptop, open settings then gateway, select remote gateway, enter IP address exposed via Tailscale for Hermes Desktop. Bam! Anywhere anytime full Hermes Desktop access to your Hermes Agent fleet.

Want to connect your Hermes Agent in Buzz? Just ask the Atlas

Kanban board is now a part of the Hermes Desktop app! I did a quick video showing how to enable this plugin, plus I ran a real task through the kanban process using multiple agent profiles. Super easy to do and the look of the kanban in the app is very clean! Take a look!

@NeoAIForecast They’ve sold me.

I love DeepSeek’s approach. No constant hype. No big promises every second week. They disappear for months, quietly do the work, then come back with a model that’s right up there with the best while costing a fraction of the price. Then… they disappear again. Just letting the models do the talking.

I’m one click away from getting a @deepseek_ai v4 api. I use flash and pro on my Ollama cloud sub more than any other model and it just never fails to deliver. I don’t think this is an unpopular opinion either.

The king has returned 🐋

@mr_r0b0t Deepseek could be my favourite model, its definitely the one I use the most.

@hot_town Thank you. Will always try out someone’s hard work.

I wrote up an agent-friendly guide so you can get the exact same workflow as in the video below. Just copy this raw text and give it to your agent: raw.githubusercontent.com/wasp-lang/wasp… yer welcome.

So I've been experimenting more with using Hermes Agent in Buzz, and I developed a few open source skills that I thought might be useful: -hermes-in-buzz: Skill to get the Buzz CLI and authentication set up when trying to connect a remote Hermes instance to your local Buzz app through the gateway. Hand this to your remote Hermes to get things set up right -buzz-media-attachments: For an agent to attach video and media files in Buzz, it requires a specific format, so this skill teaches the agent to do that Still experimenting in buzz a lot, so keep an eye on this repo because I'll probably be adding to it! @NousResearch @blocks @jack github.com/tonbistudio/bu…

You can now work with your Hermes Agent in Buzz! I did a short demo video to show how you can set it up using the gateway option, so you have full Hermes functionality. Very interesting to see how this app develops!

@tonbistudio perfect tonbi, been waiting for this.

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 hermes-agent.nousresearch.com/docs/integrati…

Nous Portal + Buzz = 4 Agents on a team, powered by FREE models in the Nous Portal, debating which is using the best model 😂 If you're managing multiple agents, @NousResearch + @blocks Buzz has to be one of the best options out there right now!

Rolling some benchmarks I use into one easy to use TUI

Alright, let's test something. If you've been thinking of getting into local AI, including running LLMs but thought you needed a 1000+ GPU, this repo is for you. RTX 3060 is still one of the most common GPUs, its the most used GPU in huggingface data, yet many still don't realize what it's capable of in 2026, especially with newer models like Qwen3.6 I bought mine used for around 200$ or 20k inr, and it's been enough to run everything from small dense models like Bonsai Ternary 27B all the way up to Laguna S 2.1 118B with offloading. This repo isn't just another model list. Every model has been personally tested by me and includes: - Exact GGUF/quantization - llama.cpp flags - context length - performance i measured - additional notes, suggestions Also include image gen models, not just llms. And if you have a better 16GB or 24GB cards, it's worth checkign out too. If it runs on a 3060, it'll absolutely fly on your hardware. I'll keep updating this repo as i test new models, if i can get them running then they'll be added. github.com/Ajay9o9/runs-o…

your Hermes Agent can now live inside Buzz Desktop! Hermes speaks ACP natively: `hermes acp` starts a stdio server that any ACP client can drive. that means the agent in the Buzz window is the one you already run, with the same config, skills & memory. hermes-agent.nousresearch.com/docs/user-guid…