I Tested the Cheapest Path to 96GB of VRAM
AMD and NVIDIA have had the obvious answers for local AI for a while... what happens when cheaper VRAM shows up from Intel. Take your personal data back with Incogni! Use code alexziskind at the link below and get 60% off an annual plan: https://incogni.com/alexziskind š Gear Links š š»ā Thunderbolt 5 external SSD: https://amzn.to/3XqetZO š»ā Favorite 15" display with magnet: https://amzn.to/3zD1DhQ š§ā” Great 40Gbps T4 enclosure: https://amzn.to/3JNwBGW š ļøš My nvme ssd: https://amzn.to/3YLEySo š¦š® My gear: https://www.amazon.com/shop/alexziskind š„ Related Videos š„ š Skip M3 Ultra & RTX 5090 for LLMs | NEW 96GB KING - Ā Ā Ā ā¢Ā SkipĀ M3Ā UltraĀ &Ā RTXĀ 5090Ā forĀ LLMsĀ |Ā NEWĀ 96...Ā Ā š» Smallest RTX Pro 6000 rig | OVERKILL - Ā Ā Ā ā¢Ā GoingĀ upĀ againstĀ Apple'sĀ UltraĀ Ā š§ Cheap mini runs a 70B LLM 𤯠- Ā Ā Ā ā¢Ā CheapĀ miniĀ runsĀ aĀ 70BĀ LLMĀ š¤ÆĀ Ā š RAM torture test on Mac - Ā Ā Ā ā¢Ā TRUTHĀ aboutĀ RAMĀ vsĀ SSDĀ UPGRADESĀ forĀ MacBookĀ Ā š FREE Local LLMs on Apple Silicon | FAST! - Ā Ā Ā ā¢Ā FREEĀ LocalĀ LLMsĀ onĀ AppleĀ SiliconĀ |Ā FAST!Ā Ā šŖ REALITY vs Appleās Memory Claims | vs RTX4090m - Ā Ā Ā ā¢Ā REALITYĀ vsĀ AppleāsĀ MemoryĀ ClaimsĀ |Ā vsĀ RTX4...Ā Ā š¦ Set up Conda - Ā Ā Ā ā¢Ā pythonĀ environmentĀ setupĀ onĀ AppleĀ SiliconĀ ...Ā Ā š¤ INSANE Machine Learning on Neural Engine - Ā Ā Ā ā¢Ā INSANEĀ MachineĀ LearningĀ onĀ NeuralĀ EngineĀ |...Ā Ā š ļø Developer productivity Playlist - Ā Ā Ā ā¢Ā DeveloperĀ ProductivityĀ Ā š AI for Coding Playlist: š - Ā Ā Ā ā¢Ā AIĀ Ā ā ā ā ā ā ā ā ā ā ā¤ļø SUBSCRIBE TO MY YOUTUBE CHANNEL šŗ Click here to subscribe: Ā Ā Ā /Ā @aziskĀ Ā ā ā ā ā ā ā ā ā ā Join this channel to get access to perks: Ā Ā Ā /Ā @aziskĀ Ā ā ā ā ā ā ā ā ā ā š± ALEX ON X: Ā Ā /Ā digitalixĀ Ā #intel #llm #b60

192GB of VRAM in One PC⦠The Cheap Way

Intel just CRUSHED Nvidia & AMD GPU pricing

AI Bubble vs Dot Com Crash. History is REPEATING

PCIe Gen 6 is... Abnormal | The Future is PCIe (Part 3), ft. Wendell of Level1 Techs

Dual Intel Arc Pro B70 GPUs⦠Worth It for AI Workflows?

Canāt get a Steam Machine? Buy this - Steamroller V2

China quietly saved the world last month

Three months wrong about why my 4-node AMD cluster was slow

I Built a $3000 AI Computer for Profit in 2026

RTX 5090, Mac Studio, or DGX Spark? I tried all three.

How do Graphics Cards Work? Exploring GPU Architecture

Recycling Destroyed Tech into Beautiful PCs

America Has a Gas Station Drug Problem... So I Tried Them

Why Google Just Gave Away Gemma 4 for Free

The (Overdue) Collapse of Bullsh*t Companies

You demanded to see this cooler... Here's how well it works!

Running a 35B AI Model on 6GB VRAM, FAST (llama.cpp Guide)

I Challenged an Employee to Build a budget AI PC⦠Big Mistake

Every Normal Thing Gen Z Got Priced Out Of (it's bad)

