PyTorch Lightning
OfficialClean training loops with built-in distributed support.
Optional ? Platforms declared: Linux, macOS, Windows
hermes skills install official/mlops/pytorch-lightning185 official skills from the Hermes Agent repository plus 101,756 community entries. Community index checked 2026-11-19 (UTC). Read community skills at their original sources.
From the Hermes Agent repository (opens in a new tab). Tree checked 2026-11-19 (UTC). Source commit 09eebd1873e9 (opens in a new tab).
Nothing in Official skills matches “”. Search the whole site
34 matching skills. Page 2 of 2. Starting points are an editorial selection, followed by name.
Clean training loops with built-in distributed support.
Optional ? Platforms declared: Linux, macOS, Windows
hermes skills install official/mlops/pytorch-lightningTrain sparse autoencoders to interpret model features.
Optional ? Platforms declared: Linux, macOS, Windows
hermes skills install official/mlops/saelensSAM: zero-shot image segmentation via points, boxes, masks.
Optional ? Platforms declared: Linux, macOS, Windows
hermes skills install official/mlops/models/segment-anything-modelvLLM: high-throughput LLM serving, OpenAI API, quantization.
Optional ? Platforms declared: Linux, macOS
hermes skills install official/mlops/inference/serving-llms-vllmReference-free preference alignment, simpler than DPO.
Optional ? Platforms declared: Linux, macOS, Windows
hermes skills install official/mlops/simpoRL post-training for LLMs with Megatron and SGLang.
Optional ? Platforms declared: Linux, macOS
hermes skills install official/mlops/slimeTeams meeting summaries, job replay, Graph subscriptions.
Bundled ? Platforms declared: Linux, macOS, Windows
Included with Hermes Agent. Read the source for setup.
High-throughput LLM inference on NVIDIA GPUs.
Optional ? Platforms declared: Linux
hermes skills install official/mlops/tensorrt-llmPretrain LLMs at scale with PyTorch 4D parallelism.
Optional ? Platforms declared: Linux, macOS
hermes skills install official/mlops/torchtitanUnsloth: 2-5x faster LoRA/QLoRA fine-tuning, less VRAM.
Optional ? Platforms declared: Linux, macOS
hermes skills install official/mlops/training/unslothOptional skills need a separate install. Some require API keys, paid services or additional software. Read the source instructions for setup. Learn how skills work.
Descriptions and declarations come from repository frontmatter. A missing platform declaration does not establish compatibility.
May we use Google Analytics to measure visits and newsletter signups? It uses cookies. Declining will not affect the site. Privacy details