This talk introduces the Sparse Abstract Machine (SAM), an abstract machine model for targeting sparse tensor algebra to reconfigurable and fixed-function spatial dataflow accelerators. SAM defines a streaming dataflow abstraction with sparse primitives that encompass a large space of scheduled tensor algebra expressions. SAM dataflow graphs naturally separate tensor formats from algorithms and are expressive enough to incorporate arbitrary iteration orderings and many hardware-specific optimizations. In this talk, we also present Custard, a compiler from a high-level language to SAM that demonstrates SAM’s usefulness as an intermediate representation. Following Custard, the SAM system also automatically binds from SAM to a streaming dataflow simulator. This talk will also provide a brief evaluation of SAM as: a general system for the whole domain of sparse tensor algebra, a design-space exploration tool for sparse accelerator performance, and as a representation that can model dataflow hardware implementations.