Acer is bringing datacenter-class AI ambitions to a compact desktop footprint with its Acer SFF RTX Spark design, showcased at IFA 2026.
Built around the NVIDIA RTX Spark superchip, the small-form-factor system is designed to run agentic AI locally while also targeting creators, developers, and gamers who need substantial compute without a traditional workstation-sized PC.
Acer says the design can deliver up to 1 petaflop of AI compute, backed by a Blackwell-based RTX GPU, NVIDIA Grace CPU, and up to 128GB of unified memory.
Acer SFF RTX Spark targets local agentic AI
The Acer SFF RTX Spark is designed around a growing PC use case: running AI models and agents directly on the user’s machine rather than relying entirely on cloud infrastructure.
Local processing can provide greater control over sensitive data while reducing the need to continuously send workloads to remote servers.
Acer is positioning the compact desktop for AI developers, creators, and gamers, giving it a broader target than a dedicated AI workstation.
The system is also designed to take advantage of NVIDIA’s full-stack AI platform and RTX technologies, allowing it to handle workloads ranging from AI development and content creation to gaming.
NVIDIA Blackwell hardware delivers up to 1 petaflop
At the center of Acer’s design is the NVIDIA RTX Spark superchip, which combines CPU and GPU resources into a unified computing platform.
The configuration showcased by Acer features up to a 6,144-core NVIDIA Blackwell RTX GPU alongside a 20-core NVIDIA Grace CPU, delivering up to 1 petaflop of AI performance. It also supports up to 128GB of unified memory in a small-form-factor desktop design.
The combination of high compute capacity and unified memory is particularly relevant to local AI workloads, where larger models can require substantial memory alongside raw processing power.
Compact desktop brings AI closer to the edge
Acer’s design reflects a broader push toward edge AI, where increasingly capable models run directly on personal computers and other local devices.
For developers, that could mean experimenting with AI agents without constantly depending on cloud computing. Creators can similarly use local AI acceleration for demanding workloads while gamers benefit from NVIDIA’s RTX ecosystem.
The compact design also makes the platform easier to fit into spaces where a conventional workstation or large desktop would be impractical.
Acer has not yet positioned the SFF RTX Spark as a shipping product, however. The company currently describes it as a design showcase for NVIDIA RTX Spark-powered systems.
Acer says availability and additional details for its RTX Spark devices will be announced at a later date. The company has not yet revealed pricing, final configurations, or Philippine availability.
