Łazzar

@digitalix I got the same with 128 ram and rtx4000 just 16Vram and get up to 88 United but sadly cannot run qwen coder next, I remember running 40t/sec but the 2 t/s and I don't remember what i messed up

i just found a $15000 lenovo laptop. wtf? and then i saw why…

@daniel_z1909 @BirantBaran No FEM means just visual slop and mockup demo for now?

@BirantBaran No FEM yet. I am working on it

⚡🤖 AI-generated visualization of an async e-motor — created only with prompts 💬 No manual modeling. Just ideas ➝ prompts ➝ physics visualization 🧲⚙️ AI turning words into engineering visuals 🔄⚡ #AI #PromptEngineering #ElectricMotor #Engineering #Simulation #Tech

@AravSrinivas @TedHZhang @perplexity_ai Make it open source maybe ;)

@TedHZhang @perplexity_ai Please let us know how we can improve it!

.@perplexity_ai Computer is fricken incredible. It has been the easiest to use agent I've used.

@Heyrohanislam Givitto me baby

EXPOSED: 200+ AI Agents have been LEAKED... These Agents SELL for $5,000+ per build and cover 20+ industries. Inside the file, you’ll get: → Social Media Agents ($8k+) → Creative Content Agents ($3k+) → Cold Outreach Agents ($3k+) → Lead Booking Agents ($6k+) → Customer Support Agents ($5k+) → Data & Scraping Agents ($8k+) → Lead Qualification Agents ($4k+) → Plus: Productivity, HR, Analytics, DevOps & Legal Agents Follow + RT + Comment “AGENTS” and I’ll send you the drive for FREE!

@Heyrohanislam AGENTS

@PeterSchiff @saylor So you use nanograms of gold 🤔

@saylor Sure there is. There are 2.1 quadrillion Satoshis. That's enough for every person on the planet to have several hundred thousand, which is several hundred thousand more than anyone needs.

There isn’t enough Bitcoin for everyone.

@sama Claude code cannot verify my number so I am stuck Codex and I am fine with that as anyway Codex is superior... even worse I use it as pluggin IN ANTIGRAVITY.... unfortunately limit are too tight even on paid plans. 5.4 my daily driver until better is available

The Codex team are hardcore builders and it really comes through in what they create. No surprise all the hardcore builders I know have switched to Codex. Usage of Codex is growing very fast:

@leeloumt @steipete I was thinking the same, just use vector but enhance with embedding 2, will report if works well

@steipete Google embedding 2 for memory not good?

There's a lot of cool stuff being built around openclaw. If the stock memory feature isn't great for you, check out the qmd memory plugin! If you are annoyed that your crustacean is forgetful after compaction, give github.com/martian-engine… a try!

@2Good2BeMe Мда, почват лекичко да се осъзнават потенциалите, вече 20 аппа от Zero to Hero, за 2 години нещата станаха ×100 по лесни и бързи, this is the way

С помощта на AI вече съм пренаписал копие на единият му продукт с 2х повече функционалност, 2х повече улеснения и много готин UI. Светът се променя.

С години сe блъсках с архаични софтуери писани от един пуяк, на който половината отговори на въпроси за нова функционалност бяха : ти знаеш ли колко работа е , това защо ти е, няма как да стане. 1/2

@TheAhmadOsman Without vram compute is NO Go , I got rtx4000 Blackwell and with 16VRam only use cas eis OCR ...

In other words: Compute is king There is a reason I won’t stop saying “Buy a GPU” P.S. If they decide to share it with us, it will be a nerfed and lobotomized version years down the road

@MichaiMorin @Agent0ai @CoeusInstitute @Docker 40 GB is massive... have you integrated llm inside it?

@MichaiMorin @Agent0ai @CoeusInstitute @Docker I'll deploy container ASAP and interested in

Push complete! We modded @Agent0ai into something better suited for our research teams. This is what happens when @CoeusInstitute gets its hands on an already impressive opensource agent OS. Now on @Docker Hub ⬇️ -- docker run michaim/coeus-mods:v1 -- Or search "michaim/coeus-mods" *Even more features added since video!* #ArtificialIntelligence #DataScience #OpenSource #automation

@MFATurkiye Whoever posted this tro Turkish x account must be illiterate 2 brain cell ameba, but anyway thanks komshu

Congratulations to our friend and ally Bulgaria on the occasion of its National Day.

@Mrcares1 @MFATurkiye Brainwashed Turkish brainrot lol... 681

@MFATurkiye Bulgaria is Turkish land, and Bulgarians are ethnically Turkish.

Your openclaw agent when you clear context window.

@Agent0ai Can we get a prompt to fix ourselves or new release is coming soon?

Agent Zero + Sonnet just did a 40-step analysis of our development vs main branch and identified a deeply hidden race condition caused by asyncio🪲. I had no idea where to start, it seemed like a websocket issue. It took A0 just 8 minutes, first try, 💯% correct.

@elonmusk They just discovered air conditioning i. Northern China so no big deal

Electricity output is the single best proxy for industrial capacity

@iotcoi @Alibaba_Qwen Lol 0.8 would run o. Smartphone let's keep the hype low

3'065.2 tok/sec generation on ZGX with the Qwen3.5-0.8B in BF16 256k context. This @Alibaba_Qwen is the BEST for AI dev and experiments!

@DopaminOg @ksimidchiyski С няколко MCPта ще е топ за егентите ми но 30т/сек си е бавно

Оказва се, че на 4 бройки Mac Studio M3 Ultra с по 512 gb ram, свързани един към друг през tb може да се run-ва Kimi 2.5 ллм модела, който е с над 1т параметри и то с 30 токена на секунда. Ше тагна @ksimidchiyski щото е топ апле фен. 😌