{"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":"Rasheed sons","offers":[{"title":"Default Title","offer_id":53661200974113,"sku":null,"price":394999.0,"currency_code":"ZAR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0739\/8958\/7233\/files\/threadripper_build_v2.png?v=1790842599","url":"https:\/\/rasheedsons.co.za\/products\/ai-workstation-pre-built-threadripper-pro-9975wx-32-core-dual-rtx-pro-4000-blackwell-48gb-128gb-ddr5-ecc-4tb-gen-5-nvme","provider":"Rasheed sons","version":"1.0","type":"link"}