Local Ai Server Benchmark 3090 vs Dual 3060s Performance is INSANE!

Checking the performance of a single 3090 vs two 3060 12GB GPUs on 5 current LLMs that you can run in a 24GB VRAM footprint. Gemma and QwQ and likely other mid 20s models get a major uplift, while smaller tens range models get an expected performance uplift.

https://digitalspaceport.com/local-ai...

$1000 Z440 AI RIG PARTS
Workstation - HP Z440 https://geni.us/HP_Z440
GPU(s) - 24GB VRAM Options
RTX 3090 24GB https://geni.us/GPU3090
RTX 3060 12GB (2x) https://geni.us/3060_GPU_12GB
SSD - 1TB EVO 870 https://geni.us/evo-870-ssd-1TB
Other HP Cheap Workstation Options
HP Z640 https://geni.us/HP_Z640
HP Z840 https://geni.us/HP_Z840

$750 Dual 3060 12GB Local Ai Server
Video    • AI Home Server 24GB VRAM $750 Budget ...  
Writeup https://digitalspaceport.com/local-ai...

Running Deepseek R1 on Z440 Local Ai Server
Video    • Deepseek R1 671b on a $500 AI PC!  
Writeup https://digitalspaceport.com/500-deep...

👍 Subscribe youtube.com/c/digitalspaceport?sub_co...

Chapters
0:00 3090 Z440 Local Ai Rig
1:00 Idle Wattage Check
2:00 QwQ 32b 3090 Benchmark
2:42 Peak Wattage Usage
3:06 Gemma 3 27B benchmark
3:48 Cogito 3090 Benchmark
4:20 Deepcoder 14B 3090 Benchmark
5:02 Gemma 3 12b benchmark
5:40 3090 Z440 Local Ai LLM Benchmark Results
6:46 Response Tokens per second
9:04 Prompt Tokens per second
12:55 used 3090s are getting cheaper

Ways to Support 🙏
🚀 Join as a member for members-only content and extra perks    / digitalspaceport  
☕ Buy Me a Coffee https://www.buymeacoffee.com/digitals...
🔳 Patreon   / digitalspaceport  

*****
As an Amazon Associate I earn from qualifying purchases.
When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
*****