Instructions to use flashrt/causal-conv1d-state with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Kernels
How to use flashrt/causal-conv1d-state with Kernels:
# !pip install kernels from kernels import get_kernel # a version (or an explicit revision) is required; see the "Files and versions" tab for the available ones kernel = get_kernel("flashrt/causal-conv1d-state", version=1) - Notebooks
- Google Colab
- Kaggle
|
Download README.md from flashrt/causal-conv1d-state: direct link, hf CLI and curl.
- Browser
- Download file 391 Bytes
-
https://huggingface.co/flashrt/causal-conv1d-state/resolve/main/README.md
- Command line
-
hf download hf://flashrt/causal-conv1d-state/README.md
-
curl -L -o README.md https://huggingface.co/flashrt/causal-conv1d-state/resolve/main/README.md
391 Bytes
metadata
library_name: kernels
tags:
- cuda
- pytorch
- flashrt
- causal-conv1d
- qwen3
- transformer
Causal Conv1D State
BF16 causal depthwise Conv1D and state-update kernels from FlashRT, packaged for Hugging Face Kernel Hub. This package is useful for transformer runtimes that keep Conv1D state on device during decode/verify/prefill.
See README.md for the public API and examples.