NVIDIA DGX Spark 64GB Gives Developers More Ways to Build and Scale Local AI
- Source
- blogs.nvidia.com
- Author
- Allen Bourgoyne
- Date

Context
NVIDIA announced a 64GB DGX Spark configuration for developers running AI models on their own machines, with partner availability scheduled for October 23. It keeps the chip and software platform of the 128GB version while providing less memory in a single unit. Announcement
The expansion path is part of the design: two 64GB units can connect by cable and pool their memory to 128GB. NVIDIA's Sync Cluster Assistant detects the machines and configures their network connection. This gives developers a route to accommodate workloads that outgrow one unit, although the announcement's performance gain of up to 1.7 times comes from NVIDIA's own test with one model and does not apply to every model or task. Configuration and test details
- The DGX Spark 64GB keeps the GB10 Grace Blackwell Superchip, DGX OS and software stack of the 128GB model, and NVIDIA says it runs models up to 100 billion parameters on device. Partner units go on sale Oct. 23 starting at $4,999.
- Two units link directly through their ConnectX-7 ports with a QSFP cable, pooling memory to 128GB for models up to 200 billion parameters. In NVIDIA's Qwen 3.8 27B test the pair delivered up to 1.7x the performance of one system.
- NVIDIA Sync Cluster Assistant detects connected units, validates their configuration and sets up the ConnectX-7 network. A Sync Model Launcher due at month end will deploy Qwen3.8 27B on one unit or a cluster and configure OpenCode to use it.
- NVIDIA says Ollama, vLLM, PyTorch with CUDA, its Toolkit and Nemotron models work out of the box, and pitches the box for always-on coding or research agents and for serving model to everyday laptops.
Checking sign-in…
Loading comments…


