BEST SELLERS

In AI & Workstation

The fans, 220 intake and 180 exhaust dual ball bearing and 40 mm thick!!!! Design (my opinion). Maybe release some custom faceplates? :) Size makes for easy building

This thing is truly amazing. Can run models up to 200 billion parameters. (Most of mine are between 70-120b, but my 24b "devstral" model is very useful for simple coding.) For a "personal AI" it's perfect. I use it to assist with sensitive client matters that I do not want to "share" with public AI services. Yes, it is brittle, but it is not hard to perform factory-restore using Nvidia's image - they have clear instructions posted and AI can help you if you need it. Be sure to use Nvidia's tool to create the boot thumb-drive. The unique hardware and driver situation makes this computer incompatible with other distros or OSes.

I had purchased in the past meshify 2 compact cases and liked them so much, I ordered this meshify 2 case, a bit larger than the compact , but my Asus Tuf Geforce Nvidia 4080 super GPU card fits this case easily. Even with the front fans. This case has plenty of options, can support 3 120/140mm fans in the front, and 1 120/140mm fan in the rear and I have a Fractal Lumen S36 cooler mounted on top. I like that this case comes apart easily giving easy access to hard to get places. Plenty of cable management and cable feed thru ports. This case is well thought out and is an excellent quality case.

It should be LEGAL LAW to use this case. I HATE ANY CASE THAT ISN'T HUGE AND GIANT.

This is pretty much as good as it gets. Yes, there are teething issues with WRX90 because it's such a new platform, but once you get past those, the board has been rock solid.

Keep in mind this card has its limitations, I am looking at it from the stand point of this verse a RTX3090 used (about the same price). No way I am spending $1000 on a used card that can die tmw even with Cuda support. So now I get 32GB vs 24GB ram. I am primary running windows and LMstudio and Ollma worked fine after installing all the intel software (like 4 large installs) Vulkan support gave me around 100T/KS for qwen 3.5-35b. Out of the box with standard setup. This running on an Intel i7-11th gen with 64GB DDR4 and PCI 4 x8 interface.. The biggest hit is on how fast I can load the model in memory. The card is very quite with the fan. I suppose other people got loud fans.

Good cable option and seems to be built solid.

- Easy setup - Super quiet - Simple remote connection options - Preloaded Jupyterlab is convenient - Playbooks always getting added - NVIDIA ecosystem in a small package

Fantastic gpu. Under like 99% usage, temps are about 60* degrees C. Can run 1440p and even 4k as long as you use x3/x4 frame gen which has little input lag. Also, it ships with ECC automatically turned on so it will say 21 gb vram. Just download intel pro graphics software and you disable it in the settings, then it should say 24.

The performance was much better than anticipated. I ran the benchmark software "stress-ng" under Rocky Linux 10.2 for twelve minutes driving all 48 threads at 100%. The highest CPU temperature was 82 degrees Celsius. The fan never exceeded 1050 RPM. I did this twice because I did not believe the first benchmark. The configuration is GIGABYTE MH53-G40 Motherboard with an AMD Ryzen™ Threadripper™ PRO 7965WX 350W SP6 - Zen 4 - 24-Core/48-Threads. There is 128 GB of Kingston FURY Renegade Pro DDR5 RDIMM and twin TEAMGROUP T-FORCE GE PRO 2TB m.2 drives with AirFlow I SSD Coolers. This is housed in a Fractal Design Define 7 XL case with 6 addidional fans on top of the three AIO fans.

- vGPU for $350 is a steal compared to Nvidia GPUs.

1) 10g nice on board 2) optional IPMI which I am planning to get. 3) Asus cpu enhancement all 24 cores fully under-load at 4.8GHz/ single core 5.2 boot stock out of the box 4) 1TB or Rdimm supported 5) lots of pcie slots for expansion