---
title: cass
canonical_url: "https://www.modelscope.cn/datasets/MBZUAI/cass"
md_url: "https://www.modelscope.cn/datasets/MBZUAI/cass.md"
repository: MBZUAI/cass
last_updated: 2025-05-28
license: "Apache License 2.0"
storage_size: "10 GB"
downloads: 653
stars: 0
---

# cass

> cass - MBZUAI 在 ModelScope 开源的数据集。💻 CASS: CUDA–AMD Assembly and Source Mapping

MBZUAI/cass 是 ModelScope 魔搭社区上的数据集，存储大小 10 GB，采用 Apache License 2.0 许可。

- **Repository**: MBZUAI/cass
- **License**: Apache License 2.0
- **Storage size**: 10 GB
- **Downloads**: 653
- **Stars**: 0
- **Last updated**: 2025-05-28

Source: https://www.modelscope.cn/datasets/MBZUAI/cass

---

# 💻 CASS: CUDA–AMD Assembly and Source Mapping

[CASS](https://huggingface.co/datasets/MBZUAI/CASS) is the **first large-scale dataset** for cross-architecture GPU transpilation, providing semantically aligned CUDA–HIP source pairs and their corresponding host/device assemblies for **NVIDIA (SASS)** and **AMD (RDNA3)** platforms. It enables research in:

* 🔁 Source-to-source translation (CUDA ↔ HIP)
* ⚙️ Assembly-level translation (SASS ↔ RDNA3)
* 🧠 LLM-guided GPU code transpilation

---

## 📚 Dataset Structure

Each sample contains the following fields:

| Field         | Description                                |
| ------------- | ------------------------------------------ |
| `filename`    | Sample ID or file name                     |
| `cuda_source` | Original CUDA source code                  |
| `cuda_host`   | Compiled x86 host-side assembly from CUDA  |
| `cuda_device` | Compiled SASS (Nvidia GPU) device assembly |
| `hip_source`  | Transpiled HIP source code (via HIPIFY)    |
| `hip_host`    | Compiled x86 host-side assembly from HIP   |
| `hip_device`  | Compiled RDNA3 (AMD GPU) device assembly   |

---

## 🔀 Dataset Splits

| Split   | Description                               | # Examples |
| ------- | ----------------------------------------- | ---------- |
| `train` | Union of `synth`, `stack`, and `opencl`   | 70,694     |
| `synth` | LLM-synthesized CUDA programs             | 40,591     |
| `stack` | Scraped and filtered CUDA from StackV2    | 24,170     |
| `bench` | 40 curated eval tasks from 16 GPU domains | 40         |

---

## 📦 How to Load

```python
from datasets import load_dataset

# 🧠 Load the full dataset (default config with all splits)
cass = load_dataset("MBZUAI/cass", name="default")

# Access a specific split
train_data = cass["train"]     # train = stack + synth + opencl
stack_data = cass["stack"]
synth_data = cass["synth"]
bench_data = cass["bench"]
```

---

## 📈 Benchmark and Evaluation

The `bench` split includes 40 samples across 16 domains like:

* 🧪 Physics Simulation
* 📊 Data Structures
* 📸 Image Processing
* 🧮 Linear Algebra

All samples have been manually verified for semantic equivalence across CUDA and HIP and come with executable device/host binaries.

---

## 📄 License

Released under the **MIT license**.

---

## 🔗 Useful Links

* 🤗 Hugging Face Collection: [CASS on Hugging Face](https://huggingface.co/collections/MBZUAI/cass-6825b5bf7414503cf16f87b2)
* 📂 Code & Tools: [GitHub Repository](https://github.com/GustavoStahl/CASS)
* Paper: [Arxiv CASS](https://arxiv.org/abs/2505.16968)