{"title":"Ai Workstation","description":null,"products":[{"product_id":"ai-workstation-pre-built-ryzen-threadripper-pro-9965wx-1284x32gb-kit-rtx-pro-4000-blackwell-24gb-gddr7-4tb-m-2-gen-4","title":"AI Workstation (Pre-built) Ryzen Threadripper Pro 9965WX, 128(4x32gb) kit, RTX PRO 4000 Blackwell 24gb GDDR7, 4TB m.2 Gen 5","description":"\u003ch3 dir=\"ltr\"\u003eRasheed Sons AI Workstation — Threadripper PRO 9965WX\u003c\/h3\u003e\n\u003cp dir=\"ltr\"\u003eA workstation built for running and fine-tuning AI models on your own hardware, under your own roof. Twenty-four Zen 5 cores, 128GB of quad-channel ECC memory and an NVIDIA RTX PRO 4000 Blackwell with 24GB of ECC GDDR7 — assembled, cable-managed and tested in Johannesburg. No cloud subscription, no per-token billing, and nothing leaving your premises.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eWhat it will actually run\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eTwo numbers decide this: 24GB of video memory, and 128GB of system memory running quad-channel at 6400MT\/s — roughly 205 GB\/s of bandwidth, which is what makes large models practical even when they don't fit on the card.\u003c\/p\u003e\n\u003cul dir=\"ltr\"\u003e\n\u003cli\u003e\n\u003cstrong\u003e7B to 14B models\u003c\/strong\u003e — Llama, Mistral, Qwen, Gemma class — run entirely on the GPU at full speed. Chat assistants, RAG over your own documents, code completion, batch document processing.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003e32B models, quantised\u003c\/strong\u003e, sit in video memory with headroom. This is where local models start genuinely competing with hosted ones for reasoning and code.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003e70B models\u003c\/strong\u003e are properly usable here. They don't fit in 24GB alone, but with 128GB of fast quad-channel memory behind the card you can split the model across both rather than being stopped by it.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eVery large mixture-of-experts models\u003c\/strong\u003e are on the table too, since 128GB holds far more than the card alone and only part of the model is active at any moment.\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp dir=\"ltr\"\u003eBeyond text: Stable Diffusion and Flux image generation, Whisper transcription, local embeddings and vector search, and LoRA fine-tuning of smaller models without renting an H100 by the hour.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eStorage that keeps up\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e4TB of PCIe Gen 5 NVMe. Model weights are large and you will collect them — a single quantised 70B is around 40GB before you have downloaded a second one. Gen 5 sequential speed means loading a model into memory takes seconds rather than minutes, which matters when you switch between them during a working day.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eBuilt to grow\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eThe graphics card is \u003cstrong\u003esingle slot\u003c\/strong\u003e, and that's deliberate. The board carries multiple PCIe Gen 5 x16 slots and the power supply is a 1600W unit, so this machine starts at one card and takes more as the workload grows — without replacing the platform. Blower cooling is what makes that work: cards exhausting out the back stay stable when stacked, where open-fan consumer cards choke on each other's hot air.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eECC on both system and video memory means a bit-flip halfway through a long run gets corrected rather than silently poisoning the result.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eSpecification\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul dir=\"ltr\"\u003e\n\u003cli\u003e\n\u003cstrong\u003eProcessor\u003c\/strong\u003e — AMD Ryzen Threadripper PRO 9965WX, 24 cores \/ 48 threads, up to 5.4 GHz\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eGraphics\u003c\/strong\u003e — NVIDIA RTX PRO 4000 Blackwell, 24GB GDDR7 ECC, 672 GB\/s, PCIe Gen 5, single slot\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eMotherboard\u003c\/strong\u003e — ASRock TRX50 WS, workstation series, PCIe Gen 5, Wi-Fi 6E\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eMemory\u003c\/strong\u003e — 128GB G.SKILL DDR5-6400 ECC Registered (4 × 32GB), quad-channel\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eStorage\u003c\/strong\u003e — 4TB XPG MARS 980 PRO, M.2 2280, PCIe Gen 5 x4, with heatsink\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eCooling\u003c\/strong\u003e — SilverStone XE360-TR5 360mm liquid cooler, full sTR5 coverage\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003ePower\u003c\/strong\u003e — ASUS Pro Workstation 1600W, fully modular\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eChassis\u003c\/strong\u003e — In Win 915 full tower, tempered glass, aluminium\u003c\/li\u003e\n\u003c\/ul\u003e","brand":"ASUS","offers":[{"title":"Default Title","offer_id":53654129967393,"sku":null,"price":348999.0,"currency_code":"ZAR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0739\/8958\/7233\/files\/screenshot-1790763541531-5.png?v=1790766518"},{"product_id":"ai-workstation-pre-built-threadripper-pro-9975wx-32-core-dual-rtx-pro-4000-blackwell-48gb-128gb-ddr5-ecc-4tb-gen-5-nvme","title":"AI Workstation (Pre-built) | Threadripper PRO 9975WX 32-Core | Dual RTX PRO 4000 Blackwell 48GB | 128GB DDR5 ECC | 4TB Gen 5 NVMe","description":"\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eRasheed Sons AI Workstation — Threadripper PRO 9975WX, Dual RTX PRO 4000\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eThe bigger brother of our 9965WX workstation. Thirty-two Zen 5 cores, 128GB of quad-channel ECC memory and two NVIDIA RTX PRO 4000 Blackwell cards for 48GB of ECC GDDR7 in total. It is assembled, cable-managed and tested in Johannesburg. Run and fine-tune AI models on your own hardware, with no cloud subscription, no per-token billing, and nothing leaving your premises.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eWhat the second card changes\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eVideo memory decides which models run at full speed. One RTX PRO 4000 gives you 24GB. Two give you 48GB, and inference tools like llama.cpp, vLLM and Ollama split a model across both cards. Models that had to spill into system memory on a single card now sit entirely on the GPUs.\u003c\/p\u003e\n\u003cul dir=\"ltr\"\u003e\n\u003cli\u003e\n\u003cstrong\u003e7B to 14B models\u003c\/strong\u003e (Llama, Mistral, Qwen, Gemma class) run on a single card at full speed, which leaves the second card free for other work.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003e32B models\u003c\/strong\u003e run with plenty of headroom for long context windows, which is what you need for RAG over large document sets and long coding sessions.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003e70B models, quantised\u003c\/strong\u003e, fit across both cards in video memory. That is the step from \"usable\" to genuinely quick, and the main reason to choose this machine over the single-card version.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eVery large mixture-of-experts models\u003c\/strong\u003e can still use the 128GB of quad-channel system memory, at roughly 205 GB\/s, alongside the 48GB on the cards.\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp dir=\"ltr\"\u003eTwo cards also means two jobs at once. Serve a chat model to your team on one GPU while the other generates images, transcribes audio or runs a fine-tune, and neither job slows the other.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eBeyond text: Stable Diffusion and Flux image generation, Whisper transcription, local embeddings and vector search, and LoRA fine-tuning without renting an H100 by the hour.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eThirty-two cores behind it\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eThe Threadripper PRO 9975WX adds eight cores over the 9965WX. That matters for everything around the model: preparing and cleaning datasets, tokenising, building vector indexes, running CPU-offloaded layers, and compiling. It also makes this a serious machine for rendering, simulation and heavy multitasking when it isn't running AI work.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eStorage that keeps up\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e4TB of PCIe Gen 5 NVMe. Model weights are large and you will collect them: a single quantised 70B is around 40GB before you download a second one. Gen 5 sequential speed means a model loads in seconds rather than minutes, which matters when you switch between models during a working day.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eBuilt for multiple cards\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eThe RTX PRO 4000 is a single-slot blower card, and that's why this works. Each card exhausts its heat out the back of the case, so two stacked cards stay stable under sustained load where open-fan consumer cards choke on each other's hot air. The 1600W power supply has ample capacity for both, and the board's remaining PCIe Gen 5 slots leave room to grow.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eECC on both system and video memory means a bit-flip halfway through a long run gets corrected instead of silently corrupting the result.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eSpecification\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul dir=\"ltr\"\u003e\n\u003cli\u003e\n\u003cstrong\u003eProcessor:\u003c\/strong\u003e AMD Ryzen Threadripper PRO 9975WX, 32 cores \/ 64 threads, up to 5.4 GHz\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eGraphics:\u003c\/strong\u003e 2 × Leadtek NVIDIA RTX PRO 4000 Blackwell, 24GB GDDR7 ECC each (48GB total), 8,960 CUDA cores and 672 GB\/s per card, PCIe Gen 5, single slot\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eMotherboard:\u003c\/strong\u003e ASRock TRX50 WS, workstation series, PCIe Gen 5, Wi-Fi\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eMemory:\u003c\/strong\u003e 128GB G.SKILL T5 NEO DDR5-6400 ECC Registered (4 × 32GB), quad-channel\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eStorage:\u003c\/strong\u003e 4TB XPG MARS 980 PRO, M.2 2280, PCIe Gen 5 x4, with heatsink\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eCooling:\u003c\/strong\u003e SilverStone XE360-TR5 360mm liquid cooler, full sTR5 coverage\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003ePower:\u003c\/strong\u003e ASUS Pro Workstation 1600W Platinum, fully modular, ATX 3.1\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eChassis:\u003c\/strong\u003e In Win 915 full tower, tempered glass, aluminium\u003c\/li\u003e\n\u003c\/ul\u003e","brand":"ASUS","offers":[{"title":"Default Title","offer_id":53661200974113,"sku":null,"price":451999.0,"currency_code":"ZAR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0739\/8958\/7233\/files\/threadripper_build_v2.png?v=1790842599"},{"product_id":"ai-workstation-pre-built-ryzen-threadripper-9980x-192gb-4x48gb-ecc-rtx-pro-5000-blackwell-72gb-gddr7-4tb-m-2-gen-5","title":"AI Workstation (Pre-built) Ryzen Threadripper 9980X, 192GB (4x48GB) ECC, RTX PRO 5000 Blackwell 72GB GDDR7, 4TB M.2 Gen 5","description":"\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eRasheed Sons AI Workstation: Threadripper 9980X, RTX PRO 5000 72GB\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eThis is our flagship AI workstation. It has sixty-four Zen 5 cores, 192GB of ECC memory and an NVIDIA RTX PRO 5000 Blackwell with 72GB of ECC GDDR7 on a single card. It is assembled, cable-managed and tested in Johannesburg. You can run and fine-tune serious AI models on your own hardware, with no cloud subscription, no per-token billing, and nothing leaving your premises.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003e72GB on one card\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eVideo memory decides what runs at full speed, and this card has 72GB of it. That's three times the RTX PRO 4000, all on a single GPU. Nothing has to be split across cards, and the whole model sits in one fast pool of memory.\u003c\/p\u003e\n\u003cul dir=\"ltr\"\u003e\n\u003cli\u003e\n\u003cstrong\u003e7B to 32B models\u003c\/strong\u003e (Llama, Mistral, Qwen, Gemma class) run at full speed with plenty of room left for long context windows, large batches or several models loaded at once.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003e70B models\u003c\/strong\u003e fit entirely on the card at higher-quality quantisation, with headroom for long context. This class of model competes properly with hosted AI for reasoning and code.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003e100B+ mixture-of-experts models\u003c\/strong\u003e such as gpt-oss-120b can sit on the card itself.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eEven larger models\u003c\/strong\u003e can use the 192GB of quad-channel system memory alongside the card, so you're not stopped by video memory alone.\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp dir=\"ltr\"\u003eBeyond text, it handles Stable Diffusion and Flux image generation at high resolution, Whisper transcription, local embeddings and vector search. It can also run LoRA and QLoRA fine-tuning of larger models without renting an H100 by the hour.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eSixty-four cores behind it\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eThe Threadripper 9980X has 64 cores and 128 threads. That speeds up everything around the model: preparing and cleaning datasets, tokenising, building vector indexes, running CPU-offloaded layers, and compiling. It's also a serious machine for rendering, simulation, virtual machines and heavy multitasking.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eStorage that keeps up\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eIt has 4TB of PCIe Gen 5 NVMe storage. Model weights are large and you will collect them: a single 70B model is 40GB or more. With Gen 5 speeds, a model loads in seconds rather than minutes, which matters when you switch between models during the day.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eBuilt to grow\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eThe RTX PRO 5000 is a dual-slot blower card. It sends its heat straight out of the back of the case, so it stays stable under long, sustained loads. The 2000W Platinum power supply and the board's spare PCIe Gen 5 slots leave room to add a second card later, without replacing the platform.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eECC on both the system and video memory means that if a bit flips halfway through a long run, it gets corrected instead of silently corrupting the result.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eSpecification\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul dir=\"ltr\"\u003e\n\u003cli\u003e\n\u003cstrong\u003eProcessor:\u003c\/strong\u003e AMD Ryzen Threadripper 9980X, 64 cores \/ 128 threads, up to 5.4 GHz\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eGraphics:\u003c\/strong\u003e PNY NVIDIA RTX PRO 5000 Blackwell, 72GB GDDR7 ECC, 384-bit, PCIe Gen 5, dual-slot blower\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eMotherboard:\u003c\/strong\u003e ASRock TRX50 WS, workstation series, PCIe Gen 5, Wi-Fi\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eMemory:\u003c\/strong\u003e 192GB G.SKILL T5 NEO DDR5-6400 ECC Registered (4 × 48GB), quad-channel\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eStorage:\u003c\/strong\u003e 4TB XPG MARS 980 PRO, M.2 2280, PCIe Gen 5 x4, with heatsink\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eCooling:\u003c\/strong\u003e SilverStone XE360-TR5 360mm liquid cooler, full sTR5 coverage\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003ePower:\u003c\/strong\u003e Cooler Master X Mighty 2000W 80 Plus Platinum, fully modular\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eChassis:\u003c\/strong\u003e In Win 915 full tower, tempered glass, aluminium\u003c\/li\u003e\n\u003c\/ul\u003e","brand":"Rasheed sons","offers":[{"title":"Default Title","offer_id":53662352867617,"sku":null,"price":484999.0,"currency_code":"ZAR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0739\/8958\/7233\/files\/threadripper_9980x_rtxpro5000.png?v=1790852793"},{"product_id":"ai-workstation-pre-built-ryzen-threadripper-pro-9985wx-256gb-4x64gb-ecc-rtx-pro-6000-blackwell-96gb-gddr7-2x-4tb-m-2-gen-5","title":"AI Workstation (Pre-built) Ryzen Threadripper Pro 9985WX, 256GB (4x64GB) ECC, RTX PRO 6000 Blackwell 96GB GDDR7, 2x 4TB M.2 Gen 5","description":"\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eRasheed Sons AI Workstation: Threadripper PRO 9985WX, RTX PRO 6000 96GB\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eThis is the top of our range. It has sixty-four Zen 5 cores, 256GB of ECC memory, and NVIDIA's flagship RTX PRO 6000 Blackwell with 96GB of ECC GDDR7 on a single card. It is assembled, cable-managed and tested in Johannesburg. This is data-centre-class AI on your own premises, with no cloud subscription, no per-token billing, and nothing leaving your building.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003e96GB on one card\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eThe RTX PRO 6000 is the largest single GPU NVIDIA makes for professional use. It has 96GB of memory on a 512-bit bus, which gives it close to 1.8 TB\/s of bandwidth. That's enough to keep the biggest models on one card, with no splitting and no compromises.\u003c\/p\u003e\n\u003cul dir=\"ltr\"\u003e\n\u003cli\u003e\n\u003cstrong\u003e70B models\u003c\/strong\u003e run at near-full precision (8-bit) entirely on the card, so you get the quality and the speed.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003e100B+ models\u003c\/strong\u003e such as gpt-oss-120b fit with plenty of room left for long context windows.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eSeveral models at once\u003c\/strong\u003e: you can serve a chat model, an embedding model and an image model side by side for a whole team.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eVery large mixture-of-experts models\u003c\/strong\u003e can use the 256GB of system memory alongside the card's 96GB. That's over 350GB of combined memory for models that won't fit anywhere else.\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp dir=\"ltr\"\u003eBeyond text, it handles Stable Diffusion and Flux image generation at full resolution, video generation, Whisper transcription, and vector search over large document sets. It can also fine-tune serious models locally: LoRA and QLoRA on 70B-class models, without renting an H100 by the hour.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eSixty-four PRO cores\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eThe Threadripper PRO 9985WX has 64 cores and 128 threads, with AMD's PRO feature set for business deployment. It handles dataset preparation, tokenising, vector indexing, CPU-offloaded layers and compilation without becoming the bottleneck. It's also a serious machine for rendering, simulation and virtual machines.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003e8TB of Gen 5 storage\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eIt has two 4TB PCIe Gen 5 NVMe drives. You can keep your operating system and tools on one drive and your model library and datasets on the other, or run them together for maximum space. A 70B model at 8-bit is around 75GB, so a serious model collection fills space quickly. With Gen 5 speeds, models load in seconds.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eBuilt for the long haul\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003eThe 2000W Platinum power supply has plenty of headroom, and the board's spare PCIe Gen 5 slots leave room to grow. ECC on both the system and video memory means that if a bit flips halfway through a multi-day training run, it gets corrected instead of silently corrupting the result.\u003c\/p\u003e\n\u003cp dir=\"ltr\"\u003e\u003cstrong\u003eSpecification\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul dir=\"ltr\"\u003e\n\u003cli\u003e\n\u003cstrong\u003eProcessor:\u003c\/strong\u003e AMD Ryzen Threadripper PRO 9985WX, 64 cores \/ 128 threads, up to 5.4 GHz\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eGraphics:\u003c\/strong\u003e PNY NVIDIA RTX PRO 6000 Blackwell, 96GB GDDR7 ECC, 512-bit, PCIe Gen 5\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eMotherboard:\u003c\/strong\u003e ASRock TRX50 WS, workstation series, PCIe Gen 5, Wi-Fi\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eMemory:\u003c\/strong\u003e 256GB G.SKILL T5 NEO DDR5-6400 ECC Registered (4 × 64GB), quad-channel\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eStorage:\u003c\/strong\u003e 8TB total: 2 × 4TB XPG MARS 980 PRO, M.2 2280, PCIe Gen 5 x4, with heatsinks\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eCooling:\u003c\/strong\u003e SilverStone XE360-TR5 360mm liquid cooler, full sTR5 coverage\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003ePower:\u003c\/strong\u003e Cooler Master X Mighty 2000W 80 Plus Platinum, fully modular\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eChassis:\u003c\/strong\u003e In Win 915 full tower, tempered glass, aluminium\u003c\/li\u003e\n\u003c\/ul\u003e","brand":"Rasheed sons","offers":[{"title":"Default Title","offer_id":53662498390305,"sku":null,"price":841999.0,"currency_code":"ZAR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0739\/8958\/7233\/files\/threadripper_9985wx_rtxpro6000_no_door.png?v=1790858628"}],"url":"https:\/\/rasheedsons.co.za\/collections\/ai-workstation.oembed","provider":"Rasheed sons","version":"1.0","type":"link"}