Acer Altos BrainSphere GB10 F1 NVIDIA GB10 AI Desktop - 128GB / 4TB
PRODUCT DESCRIPTION
Acer Altos BrainSphere GB10 F1 NVIDIA GB10 Grace Blackwell AI Desktop, 128GB Unified Memory, 4TB Self-Encrypting NVMe SSD, ConnectX-7 200GbE, Wi-Fi 7, NVIDIA DGX OS + Altos aiGeni Platform (DT.L1BSM.001) - 1 Year Onsite Local Warranty
Altos BrainSphere GB10 F1 β NVIDIA DGX Spark with a One-Click AI Platform
The Acer Altos BrainSphere GB10 F1 is Altos Computing's build of the NVIDIA DGX Spark platform: an NVIDIA GB10 Grace Blackwell Superchip pairing a 20-core Arm CPU (10 Cortex-X925 + 10 Cortex-A725) with a 6,144-CUDA-core Blackwell GPU and 128 GB of unified LPDDR5x memory, delivering up to 1 petaFLOP of FP4 AI performance in a 1.13-litre desktop under 1.5 kg. It develops, fine-tunes and runs AI models of up to 200 billion parameters locally, and two units link over ConnectX-7 to reach 405 billion.
What distinguishes the BrainSphere is the Altos aiGeni platform layered on NVIDIA DGX OS β a self-developed, one-click AI development and deployment environment that provisions PyTorch, TensorFlow, JupyterLab, Ollama, OpenWebUI, Stable Diffusion, Whisper and LLaMA Factory containers, backs up workspaces automatically and monitors CPU, GPU and memory live. SourceIT supplies the 4 TB configuration (DT.L1BSM.001) in Singapore with Acer's 1-year onsite local warranty, a full GST invoice and enterprise procurement assistance.
128 GB Unified Memory at 273 GB/s
The GB10's 128 GB of LPDDR5x is a single coherent 256-bit memory pool shared by CPU and GPU at 273 GB/s, so large language models, vision-language models and RAG pipelines load without the VRAM ceiling of a discrete-GPU workstation. Fifth-generation Tensor Cores with FP4 precision and fourth-generation RT Cores accelerate inference, fine-tuning and rendering.
4 TB Self-Encrypting NVMe, ConnectX-7 200 GbE, Wi-Fi 7
A 4 TB NVMe M.2 SSD with self-encryption holds models, datasets and containers. Networking pairs an NVIDIA ConnectX-7 NIC with two QSFP 200 Gbps ports for clustering a second unit, with a 10 GbE RJ45 port, Wi-Fi 7 and Bluetooth 5.4 LE. Three USB 3.2 Gen 2x2 Type-C ports (20 Gbps), a USB-C PD-in port (180 W EPR, PD 3.1) and an HDMI 2.1a port with multichannel audio complete the I/O.
Altos aiGeni Platform on NVIDIA DGX OS
aiGeni is a Kubernetes-based management layer for containerised GPU environments: one-click stacks for PyTorch, TensorFlow, Jupyter Lab, Ollama, OpenWebUI, Stable Diffusion, Whisper and LLaMA Factory, automatic workspace backup every 15β60 minutes, port forwarding for serving models, and live resource monitoring. Since GTC 2026 Altos also supports NVIDIA NemoClaw for agentic AI on the GB10 F1.
1.13-Litre Chassis, 170 W Maximum Draw, Under 1.5 kg
The BrainSphere GB10 F1 measures 150 Γ 150 Γ 50 mm, weighs under 1.5 kg and draws a maximum of 170 W through its USB-C PD 3.1 power input. SourceIT supplies it in Singapore with Acer's 1-year onsite local warranty.
Acer Altos BrainSphere GB10 F1 Datasheet β Technical Specifications
General
| Product Name | Acer Altos BrainSphere GB10 F1 AI Desktop |
| Part Number | DT.L1BSM.001 (4 TB configuration) |
| Category | Personal AI supercomputer / AI developer desktop (NVIDIA DGX Spark platform) |
| Form Factor | 1.13-litre mini desktop |
| Local Warranty (Singapore) | 1 Year onsite; procurement and GST invoice by SourceIT |
Processor & GPU
| Compute Platform | NVIDIA DGX Spark architecture β NVIDIA GB10 Grace Blackwell Superchip |
| CPU | 20-core Arm β 10 Γ Cortex-X925 performance cores + 10 Γ Cortex-A725 efficiency cores |
| GPU | NVIDIA Blackwell architecture GPU β fifth-generation Tensor Cores, fourth-generation RT Cores, 1 Γ NVENC, 1 Γ NVDEC |
| AI Performance | Up to 1 PFLOP (1,000 TOPS) FP4 with sparsity |
| Model Support | AI models up to 200 billion parameters on one unit; up to 405 billion parameters with two units linked over ConnectX-7 |
| CUDA Cores | 6,144, per Altos |
Memory
| Unified Memory | 128 GB LPDDR5x coherent unified system memory, 256-bit |
| Memory Bandwidth | 273 GB/s |
Storage
| SSD | 4 TB NVMe M.2 SSD with self-encryption |
Networking
| High-Speed Networking | NVIDIA ConnectX-7 NIC β 2 Γ QSFP 200 Gbps |
| Ethernet | 1 Γ RJ45 10 GbE |
| Wireless | Wi-Fi 7 + Bluetooth 5.4 with LE |
I/O Ports
| USB | 3 Γ USB 3.2 Gen 2x2 Type-C (20 Gbps) + 1 Γ USB Type-C PD in (180 W EPR, PD 3.1) |
| Display / Audio | 1 Γ HDMI 2.1a with multichannel audio output |
Operating System & Software
| Operating System | NVIDIA DGX OS (Ubuntu Linux) with NVIDIA AI software stack |
| Altos Software | Altos aiGeni one-click AI development and deployment platform; NVIDIA NemoClaw support |
Power, Physical & Environmental
| Power | USB-C PD 3.1 EPR input, 180 W; maximum power consumption 170 W |
| Dimensions (W Γ D Γ H) | 150 Γ 150 Γ 50 mm (1.13 L) |
| Weight | Under 1.5 kg |
Compatibility & Software Support
NVIDIA DGX OS and the NVIDIA AI Software Stack
Ships with NVIDIA DGX OS β an Ubuntu Linux build with the NVIDIA AI software stack preinstalled: CUDA, cuDNN, TensorRT and TensorRT-LLM, NVIDIA NeMo, NIM microservices, Triton Inference Server, RAPIDS and NGC containers, alongside PyTorch, TensorFlow, JAX, vLLM, Ollama, Hugging Face Transformers and ONNX Runtime. NVIDIA AI Workbench and the DGX Dashboard simplify environment setup, and Docker or Podman containers keep projects isolated. Because it is the same software foundation as the DGX Station and DGX data-centre systems, models developed here move to larger NVIDIA infrastructure without code changes.
Arm64 Architecture β Plan Your Software Accordingly
The GB10's Grace CPU is Arm-based, so the operating system, drivers and container images are arm64 (aarch64) builds. Mainstream AI frameworks and NVIDIA NGC containers are published for arm64, but proprietary x86-only applications will not run natively β check any commercial software you rely on before ordering. Windows is not supported on this platform.
Clustering Two Units
The two QSFP 200 GbE ports on the NVIDIA ConnectX-7 SuperNIC let two GB10 systems be linked with a QSFP112 direct-attach cable (sold separately) to work with models up to 405 billion parameters. The 10 GbE RJ45 port covers everyday site networking, and Wi-Fi 7 keeps the system connected on a desk without cabling.
Altos aiGeni Platform
aiGeni runs on top of NVIDIA DGX OS and manages containerised GPU environments through a web console: one-click deployment of PyTorch, TensorFlow, Jupyter Lab, Ollama, OpenWebUI, Stable Diffusion, Whisper and LLaMA Factory stacks, automatic workspace backup, port forwarding and live CPU/GPU/memory monitoring. It suits teams that want a managed developer experience without hand-building Docker images.
Package Contents & Ordering Information
What Ships With the System
- Acer Altos BrainSphere GB10 F1 (DT.L1BSM.001) with 4 TB self-encrypting NVMe SSD
- USB-C power adapter and power cord
- NVIDIA DGX OS with Altos aiGeni platform preinstalled
- Acer 1-year onsite local Singapore warranty
A QSFP112 direct-attach cable for linking two units (see the NVIDIA QSFP112 400G DAC cable 40 cm) is sold separately.
NVIDIA DGX Spark / GB10 Systems at SourceIT Singapore
The Acer Altos BrainSphere GB10 F1 4 TB is in stock at SourceIT Singapore with GST invoice and Acer's 1-year onsite warranty.
- NVIDIA DGX Spark 4TB β NVIDIA GB10 Grace Blackwell platform, same 128 GB unified memory and 1 PFLOP FP4 core compute
- ASUS Ascent GX10 β NVIDIA GB10 Grace Blackwell platform, same 128 GB unified memory and 1 PFLOP FP4 core compute
- Lenovo ThinkStation PGX β NVIDIA GB10 Grace Blackwell platform, same 128 GB unified memory and 1 PFLOP FP4 core compute
- Dell Pro Max with GB10 β NVIDIA GB10 Grace Blackwell platform, same 128 GB unified memory and 1 PFLOP FP4 core compute
- GIGABYTE AI TOP ATOM β NVIDIA GB10 Grace Blackwell platform, same 128 GB unified memory and 1 PFLOP FP4 core compute
- MSI EdgeXpert MS-C931 β NVIDIA GB10 Grace Blackwell platform, same 128 GB unified memory and 1 PFLOP FP4 core compute
- HP ZGX Nano G1n β NVIDIA GB10 Grace Blackwell platform, same 128 GB unified memory and 1 PFLOP FP4 core compute
- MSI XpertStation WS300 (DGX Station GB300) β step up to 748 GB coherent memory and 20 PFLOPS for trillion-parameter models
Browse the full NVIDIA DGX Spark & DGX Station collection. For volume, project, research-grant or education pricing, GST invoicing and current lead times, email sales@sourceit.com.sg.
Datasheet & Downloads
Official Acer / Altos documentation for the BrainSphere GB10 F1:
Frequently Asked Questions β Acer Altos BrainSphere GB10 F1
What is the Altos BrainSphere GB10 F1?
Altos Computing's (Acer group) build of the NVIDIA DGX Spark platform: a 1.13-litre desktop with the NVIDIA GB10 Grace Blackwell Superchip, 128 GB unified memory and up to 1 PFLOP FP4 for developing and running AI models of up to 200 billion parameters locally, with the Altos aiGeni one-click AI platform preinstalled. This configuration (DT.L1BSM.001) has a 4 TB self-encrypting NVMe SSD.
What is the Altos aiGeni platform?
Altos' self-developed AI development and deployment platform that runs on NVIDIA DGX OS: one-click container stacks for PyTorch, TensorFlow, Jupyter Lab, Ollama, OpenWebUI, Stable Diffusion, Whisper and LLaMA Factory, automatic workspace backup, port forwarding and live resource monitoring.
How does it compare to the NVIDIA DGX Station GB300 systems?
Different class: this is a desktop-scale GB10 system with 128 GB unified memory and 1 PFLOP FP4 for models up to about 200 billion parameters, aimed at individual developers, researchers and small teams. The DGX Station GB300 systems (MSI XpertStation WS300, HP ZGX Fury G1n, ASUS ET900N G3, GIGABYTE W775, Supermicro Super AI Station) carry 748 GB and up to 20 PFLOPS for trillion-parameter models and team-scale production AI.
Can I link two units together?
Yes. Two units connect over their ConnectX-7 200 GbE QSFP ports with a QSFP112 direct-attach cable (sold separately), scaling to models of up to 405 billion parameters.
What operating system does it run?
NVIDIA DGX OS, an Ubuntu Linux build with the NVIDIA AI software stack preinstalled. Windows is not supported.
Does it run Windows software or x86 applications?
No. The Grace CPU is Arm64, so only arm64 Linux software and containers run natively. Mainstream AI frameworks and NVIDIA NGC containers are available for arm64.
What warranty does it come with?
Acer's 1-year onsite local Singapore warranty, supplied through SourceIT.
How do I purchase the Altos BrainSphere GB10 F1 in Singapore?
SourceIT is an authorised IT reseller in Singapore and supplies the BrainSphere GB10 F1 with GST invoice, onsite warranty and enterprise procurement assistance β contact sales@sourceit.com.sg for volume pricing and lead times.
