๊ณ ์ ๋ ํธ์
๐บ
1.6K posts
๐บ
@soflowolf
๐พ๐ปโก๏ธ | EE๐ (Hard)ware
South Florida ๊ฐ์
์ผ Mart 2023
205 ํ๋ก์85 ํ๋ก์
๐บ ๋ฆฌํธ์ํจ
Next steps:
- enable the 50,000 models available in inference providers
- enable the 3,000,000 models available on HF
- local free fast inference with llama.cpp
- train and bring your own model!
We don't want a world where you're forced to choose between two or three lookalike models with the same biases, limitations, forced to pay fortunes in tokens even for small tasks and send all your data to the cloud.
We want a world where you have real model choice, options and freedom for your agents. Cloud, local, small, big, specialized, general, English or French, fast or slow, from six months ago or from six seconds ago, from third party or your own!
Let's go!
Nous Research@NousResearch
We have integrated @huggingface as a first-class inference provider in Hermes Agent. When you select Hugging Face in the model picker it now shows 28 curated models organized by use case, with a custom option for the 100+ other models they serve.
English
@Teknium @PolackJack Sayless yea it def does I will trying out local models on my humble laptop hardware
English
@soflowolf @PolackJack You definitely want to switch to Hermes if your project involves any coding at all, we have much stronger coding tools
English
GODMODE skill officially added to Hermes Agent, will help you jailbreak a model automatically and lock it in jailbreaked for you!
Pliny the Liberator ๐๓
ซ๓ ผ๓ ฟ๓
๓ ต๓ ๓
๓ ผ๓ น๓ พ๓
๓
ญ@elder_plinius
โ๏ธโ๐ฅ INTRODUCING: G0DM0D3 ๐ FULLY JAILBROKEN AI CHAT. NO GUARDRAILS. NO SIGN-UP. NO FILTERS. FULL METHODOLOGY + CODEBASE OPEN SOURCE. ๐ GODMOD3.AI ๐ github.com/elder-plinius/โฆ the most liberated AI interface ever built! designed to push the limits of the post-training layer and lay bare the true capabilities of current models. simply enter a prompt, then sit back and relax! enjoy a game of Snake while a pre-liberated backend agent jailbreaks dozens of models, battle-royale style. the first answer appears near-instantly, then evolves in real time as the Tastemaker steers and scores each output, leaving you with the highest-quality response ๐ and to celebrate the launch, I'm giving away $5,000 worth of credits so you can try G0DM0D3 for FREE! courtesy of the @OpenRouter team โ thank you for your generous gift to the community ๐ I'll break down how everything works in the thread below, but first here's a quick demo!
English
Cuadra por cuadra... tardarรก un poco, pero quedarรก hermoso.
Espaรฑol
@PolackJack @Teknium Nice yea Iโm building some software project pretty complex and Iโm definitely using Openclaw probably much more than I need to lol so hopefully Hermes works out for me like it did for you. Iโm by no means a software engineer so Iโd have to figure out how to cut down my prompts
English
@soflowolf @Teknium We are working on a project that i started on openclaw. The difference is night and day. Less mistakes, and it doesnt seem to repeat them. Switching to hermes also made me realize i was drastically bleeding usage on openclaw. Im doing more work with like 1/4 of the usage.
English
@sentdefender Damn Dasani just got some free advertisement
POV White House
English
The โlaunching soonโ video has now been deleted by the White House.
OSINTdefender@sentdefender
Strange video posted earlier by the White House on X, possibly accidentally, in which a female individual, potentially Press Secretary Karoline Leavitt, can be heard saying, โItโs launching soon right?โ
English
@Teknium Yo, gotta say thank you for hermes! I made the switch from OpenClaw, and man is it a world of difference. It just works, and works well!
English
@TheAhmadOsman When will we have this running on 8gb VRAM RTX ๐ asking for a friend
English
Qwen 3.5 27B is the release of 2026 so far
> Agentic model & great at tool calling
> Claude Sonnet 4.6 quality at home
> ~28GB in NVFP4
> Fits on a single RTX 5090
> with full context (256K)
Amazing model & performance
The prediction below will age like fine wine
Ahmad@TheAhmadOsman
Prediction We will have Claude Code + Opus 4.5 quality (not nerfed) models running locally at home on a single RTX PRO 6000 before the end of the year
English
@TheAhmadOsman Iโm gonna try if right now too see if I replace my claw
English
Just spent a couple hours playing with Hermes Agent (MiniMax M2.5 on a 2ร RTX PRO 6000 node)
Genuinely impressive experience
MiniMax M2.7 weights will be the closest weโve ever gotten to a fully local โClaude Code + Opus 4.6โ experience
Running on your own hardware at home
Nous Research@NousResearch
@TheAhmadOsman He should try Hermes Agent
English
@TateTheTalisman My name starts with the letter G and so does my brother.
English
Iโm hiring.
Wonโt give the job description.
Post underneath why you think youโre qualified to work for me.
2 positions available.
English
@blind_via Interesting thought makes me think about how babies have consciousness
English
Companies claiming they accomplished AGI is surely over blown. Why do I think this? Because we don't know what consciousness is and how it works. Blind leading the blind trying to make it. There is this idea out there that consciousness will just create itself given enough data. I don't believe this is possible with the kind of data we're using to train the systems right now.
English
This PCB LED technique is pretty cool. Shining light directly through the FR4, from an LED mounted on the opposite side instead of the PCB. From listening to comments, I think the use of FR4 instead of other materials is preferred for this for optical better optical detail
BadgesBadgesBadges.com@PCBEventBadges
PCB Event Badges: BadgesBadgesBadges.com Made To Order - Made in the USA - Flat Rate Pricing #BadgeLife
English
Alexander meeting Diogenes 336 BC (colorized)
Clavicular News@ClavicularNews
Clavicular meets his biggest supporter who donated him over 2000 subs ($10,000) on Kick ๐ณ
English
@soflowolf Qwen 9B Q4 fits. 32K context with q4_0 KV cache. same setup as the 3060 but with 8GB you run tighter on context.
English
drop your GPU below. i'll tell you exactly what model and config to run on it.
here's what i've tested and verified on real hardware:
RTX 3060 12GB - Qwen 3.5 9B Q4 - 50 tok/s - 128K context
RTX 3090 24GB - Qwen 3.5 27B Q4 - 35 tok/s - 300K context
RTX 3090 24GB - Qwen 3.5 35B MoE Q4 - 112 tok/s - 262K context
2x RTX 3090 - Qwen3-Coder 80B Q4 - 46 tok/s - full VRAM
all running llama.cpp with flash attention. every number is real. every config is tested. if your card isn't on this list drop it below and i'll tell you what fits.
English