Dev gets ONNX running on AMD Strix Halo by matching migraphx ABI locally
wavefnx · x · 2026-11-12
Developer wavefnx says he has been building AMD-compatible runtimes, compat layers and kernels since the start of the year while everyone piled into Nvidia. His latest demo: making Microsoft's ONNX stack work on the brand-new AMD gfx1151 (Strix Halo). Since the chip is so new that support is questionable, he had to compile ONNX locally and match its ABI against an ONNX branch with migraphx support to achieve interop. "You got front-runned," he quips — implying he positioned early while the AMD ecosystem was ignored.
More from Infra
- US Largest Public Pension Board Member Flags AI Investment Risk From Data Center Backlash — asusarla · 2026-11-12
- Agent harnesses are heading toward vertical integration, argues 0xblacklight — HankYeomans · 2026-11-12
- CoreWeave turns to customer prepayments and shorter contracts to fund AI growth — TechCrunch AI · 2026-11-12
- Foreign Investors Pull $50B From Asian Equities in October on AI Cash-Burn and Regulation Fears — nordicinst · 2026-11-12
- Bernstein Ranks Data Center Pipelines: META, MSFT, GOOGL Top the Credibility List — aronchick · 2026-11-12
- Photoroom reclaims ~1,000 idle GPU hours weekly using SkyPilot across Slurm and Kubernetes — skypilot_org · 2026-11-12