huggingface

Guides Hugging Face Hub CLI operations for downloading, uploading, managing repos, running Jobs, querying datasets, configuring Spaces, and handling Hub authentication.

Explores Hugging Face datasets through the Dataset Viewer API, resolving configs and splits, previewing rows, paginating records, searching text, filtering rows, and retrieving parquet metadata.

Runs LLM fine-tuning on Hugging Face Jobs using TRL or Unsloth, covering SFT, DPO, GRPO, reward modeling, dataset validation, hardware selection, Trackio monitoring, and Hub persistence.

Looks up Hugging Face paper pages and arXiv IDs, fetching markdown paper content and structured metadata such as authors, linked models, datasets, Spaces, GitHub repos, and project pages.

Trains and fine-tunes object detection, image classification, and SAM/SAM2 segmentation models on Hugging Face Jobs or locally with dataset validation, metrics, Trackio, and Hub persistence.

High-performance semantic caching for LLM responses using embedding similarity.

Runs Hugging Face Transformers.js models directly in JavaScript and TypeScript across browser and server runtimes for NLP, vision, audio, and multimodal inference with WebGPU or WASM.