NVIDIA Ships DGX Spark 64GB Starting at $4,999, Two Units Pool 128GB for 200B-Parameter Models
NVIDIA Blog · rss · 2026-10-02
NVIDIA announced a 64GB DGX Spark configuration shipping October 23 from Acer, ASUS, Dell, Gigabyte, HP and MSI, starting at $4,999, targeting local agent development, inference and fine-tuning.
Key points
- Keeps the GB10 Grace Blackwell Superchip and full NVIDIA software stack; runs up to 100B-parameter models on device
- Two units cluster via ConnectX-7 and Sync Cluster Assistant to pool 128GB memory, supporting up to 200B-parameter models, with up to 1.7x performance on Qwen 3.8 27B tests
- Ships with Agent Toolkit, CUDA-X, Ollama, vLLM and PyTorch; Sync Model Launcher lands end of month for one-click local model setup
- Qwen-Image-2.1 can also run locally on RTX GPUs and DGX Spark
More from Embodied
- Replacing CAD with code: LLM-written build123d Python scripts design entire robots — natesiggard · 2026-10-03
- HUST, PKU and Keenon Robotics unveil AdaRoboVLG for context-aware robot grasping — jiqizhixin · 2026-10-03
- Sharpa D01 humanoid robot outperforms humans at grasping soft dropped objects — CyberRobooo · 2026-10-03
- The $899 Mac Mini M6 was built for 24/7 agentic computing — most owners use it to check email — hey_abusiddik · 2026-10-03
- Figure AI retires F.02 humanoid fleet by training robots to leap into molten steel — heatherknight · 2026-10-03
- Meta's Muse Gadgets turns AI hardware into an open-source DIY project — The Decoder · 2026-10-03