Metraipolis › archive › dc_d291f2c81ced

ModelBlaster: Closing the Performance Gap through Learning-Augmented Agents Across Models, Kernels and Hardware Backends — arXiv:2611.04104

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.04104
License (as archived): CC0-1.0
Archived: 2026-11-04T21:51:28.868Z
SHA-256 of the archived text: 08cc77da93d8af981646ef659d0bf8908fe351af5a7a66240eb29ae519ba593d
Stable URL of this page: https://metraipolis.com/documento/dc_d291f2c81ced

How to cite

ModelBlaster: Closing the Performance Gap through Learning-Augmented Agents Across Models, Kernels and Hardware Backends — arXiv:2611.04104 — arXiv (en), https://arxiv.org/abs/2611.04104 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d291f2c81ced (Metraipolis archive dc_d291f2c81ced, sha256 08cc77da93d8af981646ef659d0bf8908fe351af5a7a66240eb29ae519ba593d)

BibTeX

@misc{dcd291f2c81ced,
  title = {ModelBlaster: Closing the Performance Gap through Learning-Augmented Agents Across Models, Kernels and Hardware Backends — arXiv:2611.04104},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.04104},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d291f2c81ced; sha256 08cc77da93d8af981646ef659d0bf8908fe351af5a7a66240eb29ae519ba593d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04104. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Optimizing accelerator code across hardware architectures is difficult. ML compilers search predefined spaces of schedules and micro-kernels, while LLM-guided optimization can explore alternative algorithms, operator decompositions, and hardware-specific techniques. However, existing LLM agents typically target a single programming interface, toolchain, or accelerator family, focus on kernel-level optimization without model-level transforms, and offer limited reuse of experience across hardware targets. We present ModelBlaster, a Memory-Augmented In-Context Reinforcement Learning (MAIC-RL) framework for optimizing data-parallel kernels and whole models across heterogeneous SIMD and SIMT accelerators. Its mutable policy is a persistent knowledge base updated at inference time by Actor-Critic Textual Gradient Descent (AC-TGD). The knowledge base maps performance states to actions, outcomes, and constraints, enabling experience reuse across kernels, models, and hardware targets. A model-level optimizer uses kernel-level knowledge to inform whole-model decisions. The same optimization loop and knowledge schema support CUDA on NVIDIA GPUs, OpenCL on Qualcomm GPUs, and RVV intrinsics on RISC-V vector cores, with only lowering and measurement layers specific to each target. Compared with PyTorch, ModelBlaster achieves geometric-mean speedups of 1.85x, 5.58x, and 5.18x on KernelBench Levels 1, 2, and 3, and 1.69x on SOL-ExecBench Levels 1 and 2 on a B200 GPU. On Qualcomm Adreno GPUs, it achieves 5.8x over the vendor SDK and 1.7x over Apache TVM; on RISC-V vector cores, it achieves 2.7x over ExecuTorch with XNNPACK across 70 KernelBench problems.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.04104 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-04T21:38:49.700Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).