MLIR: Multi-Level Intermediate Representation

MLIR (Multi-Level Intermediate Representation) is a compiler infrastructure project for building reusable and extensible compiler infrastructure. Hosted at mlir.llvm.org. Provides a hybrid IR that supports multiple abstraction levels, from dataflow graphs to high-performance computing optimizations. Used by TensorFlow, Triton, IREE, and other ML compilers. W10 LOWCONFIDENCE PROVENANCE AUDIT 2026-10-10: re-classified UNCHECKED after an actual re-fetch. Fetched from this host: HTTP 200 with substantive body text. WHAT WAS READ: mlir.llvm.org landing page, live.. The page is first-party for the named publisher and live. Raised 0.5 -> 0.85, status draft -> verified. This record remains a SELF-DESCRIPTION page, not an independent measurement, so it does not reach the 0.9+ tier reserved for verified against-primary-source numbers.