Yes—a server CPU can run some edge video-analytics inference without an installed GPU. ServeTheHome’s report from OCP Summit 2025 describes Intel demonstrating camera-based, retail-style analytics on a Dell PowerEdge XR8720t powered by an Intel Xeon 6 SoC. The report says the demonstrations ran on the CPU, with no GPU or other accelerator installed. That is evidence for a workload-specific option, not a claim that CPUs match GPUs or that every retail deployment can be sized this way.
What Intel demonstrated at OCP Summit 2025
In a December 7, 2025 event report, John Lee of ServeTheHome described an Intel edge-AI booth demonstration using cameras to simulate a retail environment. The system was a Dell PowerEdge XR8720t based on Intel Xeon 6 SoC. According to the report, multiple demonstrations ran using the CPU alone, without an installed GPU or another accelerator. ServeTheHome’s event report is the source for these observations.
As an Amazon Associate I earn from qualifying purchases.
The practical message is narrow: if an analytics model’s stream count, latency target and software stack fit the CPU, a separate GPU may not be required. The demonstration does not establish a universal architecture for stores, warehouses or other camera deployments.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11How the CPU handled the inference workload
ServeTheHome says the Xeon 6 SoC demonstrations used different CPU instruction paths. Some algorithms used Intel Vector Neural Network Instructions (VNNI) and Advanced Matrix Extensions (AMX), while others continued to use AVX2. Those capabilities can accelerate particular operations, but the report does not identify the models, camera resolutions, stream counts, batch sizes or software versions used on the show floor.
#1 Best Overall
- Dell PowerEdge R730xd 24B SFF 2U Server
- 2x Intel Xeon E5-2690 v4 2.6Ghz 14-Core (28-cores Total)
- 128GB DDR4 RAM – 4x 1.2TB 10K SAS 2.5” 12Gb/s
- Dell H730P mini 2GB 12Gb/s RAID
- 2x 750W PSU - 2x 10Gb SFP+ 2x 1Gb (RJ45) NIC
The article also states that the Xeon 6 SoC scales to up to 72 cores and includes built-in inference acceleration and a media accelerator. These are descriptions attributed to the event report, not independent measurements of this specific system.
What the Dell PowerEdge XR8720t configuration offered
ServeTheHome characterized the XR8720t as a 2U edge platform with room for two nodes and expansion options that can include GPUs. The description applies to the configuration shown at the event; it should not be treated as a complete or current specification for every XR8720t build.
Rank #2
- Renewed server with the highest quality standards
- Ideal for a robust enterprise environment or data center
- All servers include power cords, and other parts detailed in full product description below
- Custom configurations available upon request
Reported network expansion
The demonstrated configuration included Intel E830-XXVDA8F adapters. Each card was described as a dual-SFF adapter with eight 25GbE ports. Two cards would therefore provide sixteen 25GbE ports, in addition to networking provided by the Xeon 6 SoC platform itself. Confirm the exact adapter count, cabling, firmware support and port layout for any production configuration.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Why the chassis matters at the edge
A two-node-capable 2U design can consolidate processing and connectivity near cameras while retaining room for additional acceleration. That flexibility matters when a pilot begins with CPU inference but later needs GPUs, more network links or a different workload mix. The event report does not provide a complete power, cooling, storage or environmental specification.
Rank #3
- Dell 13th Generation Rack Mount 1U 8-Bay 2.5" SFF Server
- Enterprise Server For Home Use
- 2x Intel Xeon Processor E5-2690 v4 2.60GHz 14-Core CPUs
- 128GB PC4-2133 DDR4 Memory
- 2x 1TB 2.5" SATA SSDs - Solid State Drives -
What this demo proves—and what it does not
Supported conclusion
- A live, camera-based retail-like analytics scenario was shown on a Xeon 6 SoC system.
- The reported demonstrations ran without an installed GPU or other accelerator.
- CPU inference can be a viable option when the application’s performance and deployment requirements fit the platform.
Claims the report does not support
- It is not a benchmark of throughput, latency, accuracy, watts per stream or total cost.
- It does not show that CPU throughput is generally higher than GPU throughput; the report explicitly cautions against that interpretation.
- It does not establish that every retail video-analytics model, resolution or camera count will run acceptably on this CPU.
- It does not provide a vendor-validated comparison with a GPU-equipped XR8720t or another edge server.
How to decide whether CPU-only edge analytics fits
Use the OCP demonstration as a starting hypothesis, then validate your own workload. Request or measure the following before choosing a CPU-only design:
- Camera workload: Record the number of streams, resolution, frame rate, codec and whether streams are continuous or event-triggered.
- Model and pipeline: Identify the detection, classification or tracking models, preprocessing, postprocessing and framework versions. Check whether the implementation uses AVX2, VNNI, AMX or another supported path.
- Service target: Define maximum end-to-end latency, required frames per second, alert tolerances and behavior during temporary overload.
- Capacity test: Run representative video through the exact model and software stack on the proposed Xeon configuration. Measure sustained streams, tail latency, CPU utilization and memory pressure rather than relying on core count alone.
- Media handling: Verify decode and encode requirements, since camera ingest and transcoding can consume substantial resources in addition to inference.
- Deployment constraints: Check power, cooling, acoustic limits, physical space, remote management, storage retention and network redundancy at the installation site.
- Growth path: Decide whether the chassis must support a future GPU, additional nodes or more 25GbE connectivity, and confirm those options with the system vendor.
Other systems mentioned at the booth
The same booth context briefly mentioned UNEEC’s Edge-MHs system and Foxconn’s Boulder Mesa platform. The report does not provide enough information to compare their specifications, software support or performance with the Dell system, so no ranking or performance conclusion can be drawn.
Bottom line for buyers
The XR8720t demonstration is a useful counterexample to the assumption that edge video analytics always requires a GPU: Intel and Dell showed a CPU-only Xeon 6 SoC setup handling a camera-based retail simulation. Treat it as a workload-specific proof of possibility. A production decision still requires measurements using your cameras, models, latency target, network design and site constraints; the event report supplies no benchmark figures to substitute for that validation.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
