Alternatives to NVIDIA/cutile-python
LatentPagedAttention-rs
Rust + cuTile research prototype for paged latent-cache LLM decode attention, validated on an RTX 4060.
- 5
- Python
- MIT
- 9 d ago
cutile-forge
Tile-Fusion Graph compiler + hybrid PCIe-streaming LLM engine on NVIDIA CuTile (sm_89 β sm_120). 8 CUDA kernels β 3 Python tiles; 117 tok/s 7B, 70B in one 32 GB GPU at 80.8% of PCIe 5.0 ceiling; torch.compile backend.
- 0
- Python
- MIT
- 2 d ago
DeepSpeed
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
- 43k
- Python
- Apache-2.0
- 3 d ago
HarmonyOS
A curated list of awesome things related to HarmonyOS. εδΈΊιΈΏθζδ½η³»η»γ
- 20k
- C
- 2 y ago
Side by side
| Name | Stars | License | Language | Last change | In common | Compare |
|---|---|---|---|---|---|---|
| NVIDIA/cutile-python | 2.2k | None declared | Python | 7 h ago | ||
| pytorch/pytorch | 106k | None declared | Python | 1 d ago | gpu | Compare |
| SerenityOS/serenity | 34k | BSD-2-Clause | C++ | 1 d ago | kernel | Compare |
| tatavishnurao/LatentPagedAttention-rs | 5 | MIT | Python | 9 d ago | cutile, gpu | Compare |
| alacritty/alacritty | 66k | Apache-2.0 | Rust | 3 d ago | gpu | Compare |
| cilium/cilium | 26k | Apache-2.0 | Go | 18 h ago | kernel | Compare |
| samudragupto/cutile-forge | 0 | MIT | Python | 2 d ago | cutile, gpu | Compare |
| exelban/stats | 44k | MIT | Swift | 23 h ago | gpu | Compare |
| tiann/KernelSU | 21k | GPL-3.0 | Kotlin | 3 d ago | kernel | Compare |
| deepspeedai/DeepSpeed | 43k | Apache-2.0 | Python | 3 d ago | gpu | Compare |
| gvisor/gvisor | 20k | Apache-2.0 | Go | 1 d ago | kernel | Compare |
| taichi-dev/taichi | 28k | Apache-2.0 | C++ | 7 wk ago | gpu | Compare |
| Awesome-HarmonyOS/HarmonyOS | 20k | None declared | C | 2 y ago | kernel | Compare |





