# SakanaAI/sparser-faster-llms

<https://repo.masters-masters.com/en/repos/SakanaAI/sparser-faster-llms>

[Home](https://repo.masters-masters.com/en) › [SakanaAI](https://repo.masters-masters.com/en/owners/SakanaAI) › [sparser-faster-llms](https://repo.masters-masters.com/en/repos/SakanaAI/sparser-faster-llms)

SakanaAI/sparser-faster-llms has 261 stars on GitHub. Its RepoMasters card: Alpha tier. Cuda kernels for leveraging LLM sparsity to improve throughput and…

| Field | Value |
| --- | --- |
| Owner | [SakanaAI](https://repo.masters-masters.com/en/owners/SakanaAI) |
| Stars | 261 |
| Forks | 26 |
| Language | [Cuda](https://repo.masters-masters.com/en/languages/cuda) |
| License | MIT |
| Created | 2026 |
| Tier | [Alpha (250 to 699 ★)](https://repo.masters-masters.com/en/ranks/b_movie) |
| View on GitHub | [SakanaAI/sparser-faster-llms](https://github.com/SakanaAI/sparser-faster-llms) |

## Repository

Cuda kernels for leveraging LLM sparsity to improve throughput and decrease the memory requirements during inference and training.

The tier for repos with 250 to 699 ★. Published odds per card pulled: 20%. Rarity: Uncommon.

## More repos by SakanaAI

- [SakanaAI/AI-Scientist · ★ 14,940 · Jupyter Notebook · Stable](https://repo.masters-masters.com/en/repos/SakanaAI/AI-Scientist)
- [SakanaAI/AI-Scientist-v2 · ★ 7,441 · Python · Stable](https://repo.masters-masters.com/en/repos/SakanaAI/AI-Scientist-v2)
- [SakanaAI/continuous-thought-machines · ★ 2,045 · Python · RC](https://repo.masters-masters.com/en/repos/SakanaAI/continuous-thought-machines)
- [SakanaAI/ShinkaEvolve · ★ 1,496 · Python · Beta](https://repo.masters-masters.com/en/repos/SakanaAI/ShinkaEvolve)
- [SakanaAI/evolutionary-model-merge · ★ 1,441 · Python · Beta](https://repo.masters-masters.com/en/repos/SakanaAI/evolutionary-model-merge)
- [SakanaAI/text-to-lora · ★ 1,328 · Python · Beta](https://repo.masters-masters.com/en/repos/SakanaAI/text-to-lora)

## Nearby in Cuda

- [yanchi-3dv/diff-gaussian-rasterization-for-gsslam · ★ 267 · Cuda · Alpha](https://repo.masters-masters.com/en/repos/yanchi-3dv/diff-gaussian-rasterization-for-gsslam)
- [AndreSlavescu/mHC.cu · ★ 267 · Cuda · Alpha](https://repo.masters-masters.com/en/repos/AndreSlavescu/mHC.cu)
- [dnouri/cuda-convnet · ★ 261 · Cuda · Alpha](https://repo.masters-masters.com/en/repos/dnouri/cuda-convnet)
- [Dots-Infra/UltraEP · ★ 260 · Cuda · Alpha](https://repo.masters-masters.com/en/repos/Dots-Infra/UltraEP)
- [R100001/Programming-Massively-Parallel-Processors · ★ 259 · Cuda · Alpha](https://repo.masters-masters.com/en/repos/R100001/Programming-Massively-Parallel-Processors)
- [llnl/LEAP · ★ 259 · Cuda · Alpha](https://repo.masters-masters.com/en/repos/llnl/LEAP)

---
Data: GitHub · Illustration drawn by RepoMasters from the repo’s ID and its language color.