Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.10977 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917140749090816 |
|---|---|
| author | Hammond, Alec M. Markosyan, Aram Dontula, Aman Mahns, Simon Fisches, Zacharias Pedchenko, Dmitrii Muzumdar, Keyur Supper, Natacha Saroufim, Mark Isaacson, Joe Wang, Laura Hunt, Warren Gondkar, Kaustubh Levenstein, Roman Synnaeve, Gabriel Li, Richard Kahn, Jacob Mathews, Ajit |
| author_facet | Hammond, Alec M. Markosyan, Aram Dontula, Aman Mahns, Simon Fisches, Zacharias Pedchenko, Dmitrii Muzumdar, Keyur Supper, Natacha Saroufim, Mark Isaacson, Joe Wang, Laura Hunt, Warren Gondkar, Kaustubh Levenstein, Roman Synnaeve, Gabriel Li, Richard Kahn, Jacob Mathews, Ajit |
| contents | We present TritorX, an agentic AI system designed to generate functionally correct Triton PyTorch ATen kernels at scale for emerging accelerator platforms. TritorX integrates open-source large language models with a custom linter, JIT compilation, and a PyTorch OpInfo-based test harness. This pipeline is compatible with both real Meta Training and Inference Accelerator (MTIA) silicon and in hardware simulation environments for next-generation devices. In contrast to previous kernel-generation approaches that prioritize performance for a limited set of high-usage kernels, TritorX prioritizes coverage. Our system emphasizes correctness and generality across the entire operator set, including diverse data types, shapes, and argument patterns. In our experiments, TritorX successfully generated kernels and wrappers for 481 unique ATen operators that pass all corresponding PyTorch OpInfo tests (over 20,000 in total). TritorX paves the way for overnight generation of complete PyTorch ATen backends for new accelerator platforms. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_10977 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Agentic Operator Generation for ML ASICs Hammond, Alec M. Markosyan, Aram Dontula, Aman Mahns, Simon Fisches, Zacharias Pedchenko, Dmitrii Muzumdar, Keyur Supper, Natacha Saroufim, Mark Isaacson, Joe Wang, Laura Hunt, Warren Gondkar, Kaustubh Levenstein, Roman Synnaeve, Gabriel Li, Richard Kahn, Jacob Mathews, Ajit Distributed, Parallel, and Cluster Computing Hardware Architecture Programming Languages We present TritorX, an agentic AI system designed to generate functionally correct Triton PyTorch ATen kernels at scale for emerging accelerator platforms. TritorX integrates open-source large language models with a custom linter, JIT compilation, and a PyTorch OpInfo-based test harness. This pipeline is compatible with both real Meta Training and Inference Accelerator (MTIA) silicon and in hardware simulation environments for next-generation devices. In contrast to previous kernel-generation approaches that prioritize performance for a limited set of high-usage kernels, TritorX prioritizes coverage. Our system emphasizes correctness and generality across the entire operator set, including diverse data types, shapes, and argument patterns. In our experiments, TritorX successfully generated kernels and wrappers for 481 unique ATen operators that pass all corresponding PyTorch OpInfo tests (over 20,000 in total). TritorX paves the way for overnight generation of complete PyTorch ATen backends for new accelerator platforms. |
| title | Agentic Operator Generation for ML ASICs |
| topic | Distributed, Parallel, and Cluster Computing Hardware Architecture Programming Languages |
| url | https://arxiv.org/abs/2512.10977 |