Edge AI analysis pipeline running on a developer workstation

Developer hub

Build against the SDK, with engineers in the loop.

Start with the public evidence. See what is published, what is in preparation, and how to reach an engineer.

Browse resources Request an evaluation

Resource index

Know what exists before you integrate.

Public states are part of the interface. Unpublished material stays labeled until it is ready.

Documentation

In preparation

Module guides, integration notes, delivery details, and hardware setup paths will live here.

Browse SDK modules

Samples

Pending — GitHub org transfer in progress

Working capture, record, stream, and analysis pipelines will be indexed by module in the nexvdo-docs and nexvdo-samples repositories. Final public URLs are not published.

API reference

In preparation

Signatures, types, error behavior, and lifecycle notes will publish with the documentation set.

Benchmark index

Pending approval

Measured throughput and latency records will be indexed here once each figure is approved with its date, scope, and test setup.

Read the support model

See the pipeline shape first.

The public API reference is in preparation. This sketch shows the four-module flow without presenting unpublished signatures as final.

SDK modules

Included free with YUAN hardware, delivered through a supported evaluation path — engineer answer within 48 hours.
The full API surface is published with evaluation access.

pipeline.cpp

// Illustrative sketch — final signatures publish with evaluation access
auto device   = nexvdo::open_first_device();
auto session  = device.record("clip.mp4");
device.stream_preview(display_frame);
auto metadata = session.analyze(model_handle);

// Sample repositories — pending publication while the GitHub organization transfer completes.

Learning path that exists today

YUAN's Developer Hub is live. Start with Pandora, then move into the SDK.

The published path has 10 chapters. Hardware setup comes first. QCAP and QDEEP close the path with 24 NexVDO SDK tutorials.

QCAP covers capture, record, and stream. QDEEP covers analysis. This is the live evidence layer while the portal reference set is prepared.

Open YUAN Developer Hub See Pandora hardware

chapters on the published path

NexVDO SDK tutorials

QCAP

10 tutorials across capture, record, and stream.

QDEEP

14 tutorials for the analysis path.

Compatibility matrices

No inferred cells.

The table is ready for evidence. Every unpublished cell keeps its state visible.

Coverage area	Compatibility	Evidence	Artifact
Hardware targets	Matrix — pending publication	Test record — pending publication	Download — pending publication
Operating systems	Matrix — pending publication	Test record — pending publication	Download — pending publication
Module coverage	Matrix — pending publication	Test record — pending publication	Download — pending publication
Validated configurations	Matrix — pending publication	Test record — pending publication	Download — pending publication

Benchmark index — pending until approvedNo public version numbers are publishedCompatibility download — pending publication

Use it with your stack

Named integrations reflect the stacks NexVDO deployments run beside. Detailed integration notes — pending publication.

DeepStreaminference pipelinesTensorRToptimized inferenceNVENChardware encodeGPUDirectzero-copy videoJetsonedge computePyTorchmodel trainingTensorFlowmodel trainingONNXmodel interchangeOpenCVvision processingGStreamermedia pipelinesFFmpegtranscode toolingROS 2robotics middlewareC++native APIPythonscripting APIDeepStreaminference pipelinesTensorRToptimized inferenceNVENChardware encodeGPUDirectzero-copy videoJetsonedge computePyTorchmodel trainingTensorFlowmodel trainingONNXmodel interchangeOpenCVvision processingGStreamermedia pipelinesFFmpegtranscode toolingROS 2robotics middlewareC++native APIPythonscripting API

Buying a product?

You are building your own device — NexVDO SDK plus YUAN hardware inside your product.

Start at the SDK01

Commissioning a system?

You need an integrated stack delivered and supported — hardware, software, and integration through Neteon.

Request scoping02

Support model

Engineers stay in the loop.

Start with the public surface that matches the question. Every unpublished surface keeps its state visible.

Read the support model

GitHub Discussions & Issues

Ask in the open. The first response comes within 48 hours. Engineering answers remain searchable for the next team.

Read the workfloworg transfer in progress

Instant AI support

A documentation-grounded assistant is planned for direct answers from the published reference set.

“Which YUAN Edge hardware fits this pipeline?”

in preparation

Read quick answerslaunches with docs

Documentation

SDK guides, API references, sample paths, and compatibility tables will live in one index.

View the indexin preparation

Quick answers

The NexVDO documentation and API reference are in preparation. YUAN's published Developer Hub is the live learning path available today.

It is included free with YUAN hardware and delivered through a supported evaluation path. An engineer answers evaluation requests within 48 hours.

YUAN publishes 24 NexVDO SDK tutorials across QCAP and QDEEP. The nexvdo-docs and nexvdo-samples repositories remain pending while the GitHub organization transfer completes.

GitHub issues are the intended first line. Neteon handles the first response within 48 hours and escalates product questions to YUAN engineering.

Need a human?

Engineers answer evaluation requests within 48 hours. Business and hardware questions route to Neteon — YUAN’s U.S. front door.

Ask a question ↗Book a scoping call

Bring the pipeline you need to ship.

Request an evaluation. An engineer answers within 48 hours and maps the SDK, hardware, and support path to your build.

Request an evaluation

Engineer reply in 48h
Ships from U.S. stock
Free with YUAN hardware