Skip to main content

Search TrendsNew Ads Appearances Podcasts API

We value your privacy

We use cookies to understand how you use our platform and to improve your experience. Click “Accept All” to consent, or “Decline non-essential” to opt out of non-essential cookies. Read our Privacy Policy.

The Rugged Edge Media Hub

The Rugged Edge Media Hub

NVIDIA Jetson AGX Orin 32GB Super Mode on JCO-6000-ORN Series

Oct 7, 2026 · 3 min · 3 segments

During Embedded World North America, Premio Product Manager Howard Yu demonstrates NVIDIA Jetson AGX Orin 32GB running in Super Mode on Premio’s JCO-6000-ORN Series. The demo highlights how NVIDIA…

During Embedded World North America, Premio Product Manager Howard Yu demonstrates NVIDIA Jetson AGX Orin 32GB running in Super Mode on Premio’s JCO-6000-ORN Series. The demo highlights how NVIDIA JetPack 7.2 enables enhanced performance on Jetson AGX Orin while running a live VLM web UI for edge AI workloads. See how Premio’s JCO-6000-ORN Series combines NVIDIA Jetson AGX Orin with a rugged edge AI platform designed for advanced vision, inference, and real-time AI applications. Learn more about JCO-6000-ORN Series: https://premioinc.com/collections/jco-6000-orn-series-ai-edge-computer

During Embedded World North America, Premio Product Manager Howard Yu demonstrates NVIDIA Jetson AGX Orin 32GB running in Super Mode on Premio’s JCO-6000-ORN Series. The demo highlights how NVIDIA…

Speakers: Howard YuGuest

Top entities: Nvidia×7 Vision-language model×4 web user interface Artificial intelligence

VLM Demo and JetPack 7.2 Performance0:22 – 2:30

H

0:22

And it's running with the JPEG 7.2. What's the JPEG 7.2 release benefit for 32GB? It provides a super mobile app.

H

0:33

It increases the tops from 200 to 240.

H

0:39

It's more higher performance compared to the original 32GB.

H

0:47

And now we're trying to run the demo with the VLM Live.

H

0:52

H

0:54

What does VLM stand for? Vision Language Model.

H

0:59

So what is wrong with it? We use the camera to see everything.

H

1:04

And then after that, we use the model to describe anything it sees.

H

1:08

Not only describe, you can ask any questions.

H

1:12

You can do the object detection.

H

1:14

You can even try to ask him something and ask him to give you some alarm.

H

1:21

So this demo, we're currently running on the Super mode.

H

1:24

And then this Super mode makes it run with the average around 38-90 ms for one of the prompts.

H

1:33

And then what we can see is the comparison between old 40W and the Super mode.

H

1:39

Let's see.

H

1:40

Now I switch to the 40W and you can see the latency.

H

1:45

Right now it's going higher and higher.

H

1:47

It takes more time to run one of the prompts.

H

1:50

So what's the difference? The Super mode lets you do more high performance and spend less time to run the model.

H

2:00

But the original 40W mode, it takes more time to run it.

H

2:08

This is how, why we're saying JPEG 7.2 is really benefit for 32GB.

H

2:15

It's higher and it's more AI performance and then you don't have to spend more money on the 64GB on the 8GX.

H

2:25

So this is how our JCO 6000 we release with the 7.2.

NvidiaCompanyCompany
88%7 mentions
Vision-language modelEntity
63%4 mentions
web user interfaceEntity
38%1 mention
Artificial intelligenceEntityProfessional field
13%1 mention

AI HardwareTechnology > Computing
Edge AI PlatformsTechnology > Computing > AI Hardware
JCO 6000 32GBTechnology > Computing > AI Hardware > Edge AI Platforms
JPEG 7.2 ReleaseTechnology > Software > AI Software Optimization
Super Mode PerformanceTechnology > Software > AI Software Optimization > JPEG 7.2 Release
Vision Language ModelsTechnology > Artificial Intelligence > Multimodal AI
VLM Live Web UITechnology > Artificial Intelligence > Multimodal AI > Vision Language Models
Real-Time Vision SystemsTechnology > Computer Vision
Camera-Based Scene UnderstandingTechnology > Computer Vision > Real-Time Vision Systems
Object Detection and AlertsTechnology > Computer Vision > Real-Time Vision Systems > Camera-Based Scene Understanding
AI InferenceEngineering > Performance Engineering
Latency BenchmarkingEngineering > Performance Engineering > AI Inference
40W Mode Versus Super ModeEngineering > Performance Engineering > AI Inference > Latency Benchmarking
Computer VisionTechnology
Multimodal AITechnology > Artificial Intelligence
AI Software OptimizationTechnology > Software
Technology
ComputingTechnology
Artificial IntelligenceTechnology
Engineering
Performance EngineeringEngineering
SoftwareTechnology

VLM Demo and JetPack 7.2 Performance0:22 – 2:30

H

0:22

And it's running with the JPEG 7.2. What's the JPEG 7.2 release benefit for 32GB? It provides a super mobile app.

H

0:33

It increases the tops from 200 to 240.

H

0:39

It's more higher performance compared to the original 32GB.

H

0:47

And now we're trying to run the demo with the VLM Live.

H

0:52

H

0:54

What does VLM stand for? Vision Language Model.

H

0:59

So what is wrong with it? We use the camera to see everything.

H

1:04

And then after that, we use the model to describe anything it sees.

H

1:08

Not only describe, you can ask any questions.

H

1:12

You can do the object detection.

H

1:14

You can even try to ask him something and ask him to give you some alarm.

H

1:21

So this demo, we're currently running on the Super mode.

H

1:24

And then this Super mode makes it run with the average around 38-90 ms for one of the prompts.

H

1:33

And then what we can see is the comparison between old 40W and the Super mode.

H

1:39

Let's see.

H

1:40

Now I switch to the 40W and you can see the latency.

H

1:45

Right now it's going higher and higher.

H

1:47

It takes more time to run one of the prompts.

H

1:50

So what's the difference? The Super mode lets you do more high performance and spend less time to run the model.

H

2:00

But the original 40W mode, it takes more time to run it.

H

2:08

This is how, why we're saying JPEG 7.2 is really benefit for 32GB.

H

2:15

It's higher and it's more AI performance and then you don't have to spend more money on the 64GB on the 8GX.

H

2:25

So this is how our JCO 6000 we release with the 7.2.

Sign up to see more

The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode

Create an account Already have an account? Log in

No account is needed to search Radar.