Just some numbers so you don’t get misled RTX 3090 (7 years old)
> 24GB VRAM
> Bandwidth: 936.2 GB/s
> Bi-directional NVLink 112GB/s RTX PRO 4000
> 24GB VRAM
> Bandwidth: 672 GB/s > No Bi-directional NVLink,
> need 32 Gen. 5 PCIe Lanes to pool 2 at 64GB/s x.com/LLMJunky/statu…
@theahmadosman
-
RTX PRO 4000 vs RTX 3090 VRAM Bandwidth and NVLink Compared
By
–
-
Run LLMs Locally on Your Own Hardware With RTX 3090s
By
–
re: Anthropic, Dario, OpenAI, etc Don’t let them control your Intelligence Utilization It is a MUST that you learn how to run your LLMs locally on your own hardware 2x RTX 3090s and Qwen 3.6 27B is all you need to get started
-
Open Source AI Is the Only Defense Against Corporate Control
By
–
Yup They keep trying to lobby to ban opensource AI because that’s the only thing standing in their way; people having the option and the freedom to run models that they don’t control They CANNOT be trusted
-
Remote Infrastructure Management via SSH WireGuard DNS and GPU Job Dispatch
By
–
~2,000 miles from home – SSH into the mothership
– WireGuard tunnel like I’m on LAN
– Resolve everything over private DNS
– Routes via reverse proxies – Dispatch jobs to GPUs
– Sync state across agents
– Search conversations, traces, artifacts, logs Read how below -
Tenstorrent Launches TT-Deploy for AI Hardware Deployment
By
–
It's TT-Deploy day Really excited to see what folks at Tenstorrent have been cooking
-
In-Depth Overview Replicable via Codex or AI Agent
By
–
Many of you asked for an in-depth overview of this Here it is, if you send this article to Codex or any agent, it can help you replicate it
-
GPU Rental Market Worsens: Availability and Interconnect Opacity
By
–
The quoted tweet is about API endpoints, but the same thing applies to hardware you don’t control Renting GPUs is not what it was a year ago, availability is so much worse And even when you find capacity, you often have no real idea what the interconnect actually looks like
-
Homelab Becomes Personal Cloud With GPU and AI Agents
By
–
~2,000 miles from home – SSH into the mothership
– WireGuard tunnel like I’m on LAN
– Resolve everything over private DNS
– Routes via reverse proxies – Dispatch jobs to GPUs
– Sync state across agents
– Search conversations, traces, artifacts, logs My homelab is now the cloud -
Bay Area meetup: GPUs, local AI, and inference engines discussion
By
–
In the Bay Area for the next couple of weeks If you’re around and wanna grab food, coffee, or yap about GPUs / local AI / inference engines / the future of owning the stack Hit me up
-
Tech Giant’s Anti-Open Source Practices and Ethical Concerns
By
–
Bro, you’re wasting your time They’re evil through and through; lobbying against opensource, rug pulling their users, fear mongering 24/7, and lying all the time Ultimately, Amazon is going to acquire them for parts and I believe humanity is better off that way