All skills
flash-attention
Speeds up training and inference for long-sequence transformer models.
- Part of NousResearch/hermes-agent
- 7 installs
- Machine Learning · GPU acceleration
More GPU acceleration skills
See all- cudaq-guide2.5kGuides CUDA-Q setup, simulation targets, QPU access, and writing quantum kernels.
- agent-matrix-optimizer1.1kOptimizes matrix computations used in numerical and machine learning workloads.
- compute-sanitizer-diagnose-cuda-bugs105Finds CUDA memory errors, data races, uninitialized reads, and sync bugs and explains how to read the report.
- cluster-network-analysisEstimates cross-server traffic, network bottlenecks and sync waits for multi-node training or inference.
Also in this pack
- hermes-agent1.4kConfigures, themes, extends, and orchestrates Hermes Agent itself.
- humanizer1.4kRewrites text to remove AI-sounding phrasing and add a natural voice.
- popular-web-designs1kProvides 54 real design systems like Stripe, Linear and Vercel as HTML and CSS.
- arxiv763Searches arXiv papers by keyword, author, category or ID.