Remote access

Try FastFlowLM without a Ryzen™ AI laptop

Launch our hosted FastFlowLM + Open WebUI environment backed by an AMD Ryzen™ AI box with 96 GB RAM. Stream tokens, upload images, or run agentic workflows exactly as you would locally.

Credentials

Username: guest@flm.npu
Password: 0000

Prefer isolation? Create your own account inside Open WebUI. Sessions reset nightly.

What's included

Preloaded models & tooling

  • Models

    Llama 3.2 3B/8B, Gemma 3 4B Vision, DeepSeek-R1 7B, Whisper, EmbeddingGemma.

  • Integrations

    Open WebUI, REST playground, LangChain quickstart scripts.

  • Monitoring

    Live NPU telemetry so you can watch throughput and power draw.

Limits

  • Context

    4,096 tokens per chat for fairness.

  • Availability

    Server may reboot for OS updates. Try again in a few minutes if offline.

Bring your team

Need a guided walkthrough?

Schedule a live session with the FastFlowLM team. We’ll tailor the demo to your use case (assistants, RAG, multimodal) and share logs afterward.