# CASS: Nvidia to AMD Transpilation with Data, Models, and Benchmark

arXiv: https://arxiv.org/abs/2505.16968 · HTML: https://mlsanity.com/paper/2505.16968/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 43 references that are in the map.
- First posted on arXiv 22 May 2025.

## Builds on (most cited first, 9 of 9)

- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [StarCoder 2 and The Stack v2: The Next Generation](https://mlsanity.com/paper/2402.19173/) (2024)
- [Liger Kernel: Efficient Triton Kernels for LLM Training](https://mlsanity.com/paper/2410.10989/) (2024)
- [Testing GPU Numerics: Finding Numerical Differences Between NVIDIA and AMD GPUs](https://mlsanity.com/paper/2410.09172/) (2024)
- [Guess & Sketch: Language Model Guided Transpilation](https://mlsanity.com/paper/2309.14396/) (2023)
- [From CISC to RISC: language-model guided assembly transpilation](https://mlsanity.com/paper/2411.16341/) (2024)

## Built on by (most cited first, 2 of 2)

- [AMDKernelVault: Large-Scale Datasets and Agentic Training for AMD GPU Kernel Optimization](https://mlsanity.com/paper/2609.12471/) (2026)
- [MetaKernelBench: Measuring GPU Kernel Knowledge Transfer Beyond Code](https://mlsanity.com/paper/2610.05014/) (2026)

## Abstract (arXiv)

Cross-architecture GPU code transpilation is essential for unlocking low-level hardware portability, yet no scalable solution exists. We introduce CASS, the first dataset and model suite for source- and assembly-level GPU translation (CUDA <--> HIP, SASS <--> RDNA3). CASS contains 60k verified host-device code pairs, enabling learning-based translation across both ISA and runtime boundaries. We generate each sample using our automated pipeline that scrapes, translates, compiles, and aligns GPU programs across vendor stacks. Leveraging CASS, we train a suite of domain-specific translation models that achieve 88.2% accuracy on CUDA -> HIP and 69.1% on SASS -> RDNA3, outperforming commercial baselines including GPT-5.1, Claude-4.5, and Hipify by wide margins. Generated code matches native performance in 85% of cases, preserving both runtime and memory behavior. To support rigorous evaluation, we introduce CASS-Bench, a curated benchmark spanning 18 GPU domains with ground-truth execution. All data, models, and evaluation tools will be released as open source to support progress in GPU compiler tooling, binary compatibility, and LLM-guided code translation.