InfraSight AI Platform

Discover AI Solutions

Browse applications and models, choose execution engines and backends, and follow step-by-step setup to run experiments on AMD hardware or in the cloud.

Choose Your Path

Explore

Explore Applications

Explore AI applications across Smart Infra, Machine Vision, and Smart Retail verticals with ready-to-deploy models.

Explore

Explore Models

Find the right AI model for your use case with our interactive model comparison and selection tool.

Coming Soon

Sensor-to-AI Data Streaming Pipeline

Reference design for deterministic sensor ingest with on-device AI acceleration.

Coming Soon

Convert CUDA to HIP

Migrate your CUDA applications to HIP with documentation, porting guidance, and structured workflows.

Go in steps

Pick a use case, choose an application, decide where it runs, then benchmark models, all in one guided flow.

Versatile Models

Models across diverse modalities, each with device-level performance data so you can compare before you commit.

Vision Detection, tracking & inspection
Language On-device LLMs & chat
Audio TTS & spoken interfaces
Multimodal Vision-language understanding

On Your Hardware or AMD Cloud

Run locally with full privacy or evaluate on AMD cloud - same walkthroughs, same setup experience.

On Your Device

Deploy and run AI models directly on your AMD hardware—NPU, iGPU, or CPU for maximum privacy, security, and local performance.

or

AMD Cloud

Remotely access the same AMD AI hardware in the cloud for prototyping and benchmarking models.

With Multi-Target Support

Swap targets in the model view to compare engines, throughput, and accuracy across Ryzen AI and embedded SKUs.

Strix Halo
Ryzen AI Max Series
Overall PerformanceUp to 126 TOPS
CPU CoresUp to 16C / 32T
iGPURadeon 8060S
Best ForHigh-perf AI
Strix Point
Ryzen AI 300 Series
Overall PerformanceUp to 85 TOPS
CPU CoresUp to 12C / 24T
iGPURadeon 890M / 880M
Best ForMainstream Edge AI
R8000
Embedded Series
Overall PerformanceUp to 39 TOPS
CPU CoresUp to 8C / 16T
iGPURadeon 780M / 760M / 740M
Best ForEmbedded & IoT Edge