Abstract
Sparse-accelerator design spaces are usually searched against analytical models, so a design point is admitted on what a model predicts rather than on what the hardware does. SparseCraft closes that gap with a language model inside a closed CHIA loop. In each of 15 iterations the model reads the measured outcome of the previous one and edits the Chisel RTL, the memory configuration and the sparse-kernel schedule of a Gemmini accelerator through MCP tool servers, and no candidate counts until it has been checked for legality, elaborated, simulated cycle-accurately, checked bit-for-bit on every output against a golden reference, and synthesised. The harness turns each measurement into the next work order, a diagnosed bottleneck with matching strategy guidance, the history of tried designs and a score of the model's own prediction, and a second model repairs changes that fail a gate. On a $512 \times 512$ GraphChallenge sparse-DNN layer the loop reaches 2.1x fewer cycles, 9.8x less off-chip traffic and 22.8% less area than the block-sparse Gemmini baseline, with 5.61x higher modelled perf/W and 11.8x lower EDP. The levers span three layers: a schedule that keeps the dense operand resident removes 9.8x of the traffic, a zero-gated MAC and a zero-row skip unit that the model wrote in Chisel cut energy, and resizing the memories cuts area.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Chakraborty, R., Samartha, M. P., Pahariya, V., & Shukla, P. (2026). SparseCraft: Agentic Hardware-Software Co-Optimization for Sparse Computing. https://omanscience.com/en/articles/sparsecraft-agentic-hardware-software-co-optimization-for-sparse-computing
MLA 9
Chakraborty, Rajatabha, et al. "SparseCraft: Agentic Hardware-Software Co-Optimization for Sparse Computing." https://omanscience.com/en/articles/sparsecraft-agentic-hardware-software-co-optimization-for-sparse-computing.
Chicago (author–date)
Chakraborty, Rajatabha, M P Samartha, Vedant Pahariya, and Priyesh Shukla. 2026. "SparseCraft: Agentic Hardware-Software Co-Optimization for Sparse Computing." https://omanscience.com/en/articles/sparsecraft-agentic-hardware-software-co-optimization-for-sparse-computing.
Harvard
Chakraborty, R., Samartha, M. P., Pahariya, V. and Shukla, P. (2026) 'SparseCraft: Agentic Hardware-Software Co-Optimization for Sparse Computing', Available at: https://omanscience.com/en/articles/sparsecraft-agentic-hardware-software-co-optimization-for-sparse-computing.
Vancouver
Chakraborty R, Samartha MP, Pahariya V, Shukla P. SparseCraft: Agentic Hardware-Software Co-Optimization for Sparse Computing. https://omanscience.com/en/articles/sparsecraft-agentic-hardware-software-co-optimization-for-sparse-computing
IEEE
R. Chakraborty, M. P. Samartha, V. Pahariya, and P. Shukla, "SparseCraft: Agentic Hardware-Software Co-Optimization for Sparse Computing," https://omanscience.com/en/articles/sparsecraft-agentic-hardware-software-co-optimization-for-sparse-computing.