Docs & Recipes
Improve instructions and guides
Submit docs PRs, fix typos, or add your own notebooks under docs/instructions.
Community
A developer-first OSS project
FastFlowLM is shaped by laptop modders, enterprise AI teams, and AMD engineers who meet in Discord, GitHub issues, and weekly office hours. Bring your benchmark, integration idea, or feature request.
Weekly cadence
-
Patch review hour
Fridays @ 9am PT · Live stream in Discord
-
Office hours
One-on-one help with NPUs, models, or deployment hurdles.
-
Community demos
Show off what you’re building on Ryzen™ AI and get feedback.
Ways to contribute
Pick the path that matches your stack
Runtime
Optimize kernels + schedulers
Help us squeeze more throughput out of XDNA2 with fused ops and better memory planning.
Integrations
Bring FastFlowLM into your stack
Contribute connector templates for LangChain, AutoGen, Open WebUI, or your product.
Code of conduct
Be kind, stay curious, keep shipping
We want a welcoming space where seasoned kernel engineers and new builders collaborate. Respect fellow contributors, share repro steps, and celebrate wins loudly.
Need help?
Ping @fastflowlm in Discord or email info@fastflowlm.com.
We respond within 24 hours.