---
title: prime-convergents-cuda
canonical_url: "https://www.modelscope.cn/models/cahlen/prime-convergents-cuda"
md_url: "https://www.modelscope.cn/models/cahlen/prime-convergents-cuda.md"
repository: cahlen/prime-convergents-cuda
last_updated: 2026-04-15
license: mit
library_name:
  - pytorch
frameworks:
  - pytorch
downloads: 3
stars: 0
tags:
  - kernels
  - cuda
  - prime-convergents
  - continued-fractions
  - erdos-mahler
  - number-theory
---

# prime-convergents-cuda

> prime-convergents-cuda - cahlen 在 ModelScope 开源的模型。Prime Convergents of Continued Fractions (v2, uint128)

cahlen/prime-convergents-cuda 是 ModelScope 魔搭社区上的机器学习模型，采用 mit 许可。

- **Repository**: cahlen/prime-convergents-cuda
- **License**: mit
- **Tags**: kernels, cuda, prime-convergents, continued-fractions, erdos-mahler, number-theory
- **Downloads**: 3
- **Stars**: 0
- **Last updated**: 2026-04-15

Source: https://www.modelscope.cn/models/cahlen/prime-convergents-cuda

---

# Prime Convergents of Continued Fractions (v2, uint128)

Computes prime statistics of CF convergents using uint128 arithmetic (depth ~75). Tracks primality, doubly-prime convergents, and Erdos-Mahler bound.

## Usage

```python
import torch
from kernels import get_kernel

kernel = get_kernel("cahlen/prime-convergents-cuda")
result = prime_conv.compute(num_samples=1000000, max_depth=500, mode=0)
```

## Compile (standalone)

```bash
nvcc -O3 -arch=sm_90 -o prime_convergents prime_convergents/prime_convergents_v2.cu -lm
```

## Results

All computation results are open:
- **Website**: [bigcompute.science](https://bigcompute.science)
- **Datasets**: [huggingface.co/cahlen](https://huggingface.co/cahlen)
- **Source**: [github.com/cahlen/idontknow](https://github.com/cahlen/idontknow)

## Citation

```bibtex
@misc{humphreys2026bigcompute,
  author = {Humphreys, Cahlen},
  title = {bigcompute.science: GPU-Accelerated Computational Mathematics},
  year = {2026},
  url = {https://bigcompute.science}
}
```

*Human-AI collaborative. Not peer-reviewed. All code and data open.*