India's value-for-money IT hardware brand
14 years. 28,000 sq ft. 100,000+ devices. From desktops to data center — designed, engineered, and manufactured in India.
Why choose RDPDesktops to data center, all Make in India
14 product categories across compute, AI, and data center. Deployment-ready from our 28,000 sq ft facility.
Download Product CatalogSovereign AI infrastructure
End-to-end AI compute under one sovereign umbrella. Designed here. Manufactured here. Supported here.
Talk to a Solutions Architect
SLA-driven. Not ticket-driven.
Warranty. SLA. On-site service. Account management. Every commitment documented, every response time defined.
Download SLA Commitment
Built on process, not promises
ISO 9001. PLI 2.0. SOP-led manufacturing. The systems behind every device we ship.
Our Story
AR-950
Halo compact AI workstation on AMD Ryzen AI Max+ 395 with Radeon 8060S & 128GB unified LPDDR5X — for high-memory local AI, multimodal demos & private model appliances.
Key Applications
- High-Memory Local AI Workstation
- Local Multimodal Model Demo
- Private Model Appliance
- Premium ISV Solution Box
- AI Founder Workstation
- Edge AI Node
- Local LLM Serving
- Developer AI
Key Features
- AMD Ryzen AI Max+ 395 class with integrated NPU and large Radeon 8060S iGPU — halo AI platform
- 128GB unified LPDDR5X — keep large local models resident in memory
- M.2 2280 PCIe 4.0 — 2TB NVMe (higher project options) for model libraries
- AMD Radeon 8060S graphics · up to 4 independent displays
- Dual 2.5GbE + USB4 + PCIe expansion in a compact workstation chassis
- Advanced AI workstation image: local serving, vector DB, agent orchestration & developer stack · Make in India
Product SKU
| SKU | AR-950 |
|---|---|
| Family | RDP A Series — AR (Agentic Ryzen) |
| Tier | Halo · 950 (AMD Ryzen AI Max class) |
| Positioning | Halo compact AI workstation |
Processor
| CPU |
|
|---|---|
| Architecture | AMD Zen 5 (Ryzen AI Max+) with XDNA 2 NPU and large integrated Radeon graphics |
| Socket | Soldered BGA (mobile-class), single processor |
Final core/thread counts and clocks depend on the selected Ryzen AI Max+ SKU — to be validated against the AMD datasheet before publication.
AI Acceleration
| NPU | Integrated AMD Ryzen AI NPU (XDNA 2) for local inference, assistants & agents |
|---|---|
| GPU compute | Large AMD Radeon 8060S iGPU for GPU-accelerated local model inference |
| Platform AI | Combined CPU + NPU + large Radeon 8060S iGPU with 128GB unified memory |
| AI runtime | Optional validated image: local model runtime, vector database & container runtime |
NPU TOPS, memory bandwidth and platform AI performance figures to be validated against final AMD datasheets and RDP engineering certification.
System Memory
| Configured |
|
|---|---|
| Maximum | 128GB (unified) |
| Type | LPDDR5X unified, soldered on-package |
| Slots | Soldered — no SO-DIMM (high-bandwidth unified memory) |
Storage
| Primary Drive | 2TB M.2 NVMe (minimum) · higher project options |
|---|---|
| Slots | M.2 2280 PCIe 4.0 SSD |
| Expansion | Project storage options for large model libraries |
Graphics & Display
| Graphics | AMD Radeon 8060S graphics |
|---|---|
| Video Output | 1× HDMI 2.1 · 1× DP 2.1 · 1× USB-C (USB4 DP-Alt) |
| Multi-Display | Up to 4 independent displays (quad display) |
Networking & I/O
| Onboard LAN | 2× 2.5GbE RJ45 |
|---|---|
| USB | 3× USB 3.2 + USB-C · 1× USB-C (USB4, 40Gbps) |
| Expansion | USB4 + PCIe expansion (Ryzen AI Max+ platform) |
| Audio | 1× 3.5mm combo audio jack |
Wireless
| Wi-Fi | Wi-Fi 6E |
|---|---|
| Bluetooth | Bluetooth 5.3 |
Chassis & Physical
| Form Factor | Compact AI workstation (NUC-class+) |
|---|---|
| Dimensions | Compact workstation chassis — final dimensions per industrial design |
| Mounting | Desk / VESA mount (project option) |
Power
| Adapter | External DC adapter — higher-wattage for Ryzen AI Max+ platform |
|---|---|
| Input | 100–240V AC, 50/60Hz |
Software & AI Stack
| OS Image | Advanced AI workstation image (local serving, vector DB, agent orchestration, developer stack) |
|---|---|
| AI Runtime | Optional: local model runtime, vector database, container runtime, agent orchestration |
| Agent Stack | OpenClaw / Hermes validation target (optional factory image, post engineering certification) |
| Management | Remote management option · RDP QC benchmark report |
OS Support, Compliance & Warranty
| Operating System | Ubuntu LTS / Linux AI workstation image · Windows 11 Pro option |
|---|---|
| Compliance | CE / FCC / BIS — per final product certification |
| Warranty | India warranty & support — final SLA confirmed at launch |
| Manufacturing | Make in India — RDP-led productization & support, Hyderabad |
Final CPU availability, OS images, certification, warranty/SLA, and commercial configurations may vary by project and procurement requirement.
Request a quote for AR-950
Tell us your AI workload — high-memory local AI, multimodal demo, or private model appliance — along with target quantity and timeline. Our team will spec the right configuration, OS/AI image, and pilot plan.
Pilot programs: available in 5–25 unit quantities for enterprise pilots, SI deployments, and ISV appliance bundles.