HuggingFace
certifiedaiLocal HuggingFace artifact consumption — pull, pipeline, embed, tokenize, dataset, cache
pip install chp-adapter-huggingfaceAuto-discovered through the chp.adapters entry point — no host config needed.
Capabilities
What this adapter exposes.
apply_adapterv1.0.0chp.adapters.huggingface.apply_adaptercache_infov1.0.0Scan the local HuggingFace cache and return a storage summary by artifact for governance and quota management.
chp.adapters.huggingface.cache_infocall_spacev1.0.0chp.adapters.huggingface.call_spaceclassify_imagev1.0.0chp.adapters.huggingface.classify_imagedataset_previewv1.0.0chp.adapters.huggingface.dataset_previewembedv1.0.0chp.adapters.huggingface.embedevaluatev1.0.0chp.adapters.huggingface.evaluatefaiss_indexv1.0.0chp.adapters.huggingface.faiss_indexfinetunev1.0.0chp.adapters.huggingface.finetunegenerate_imagev1.0.0chp.adapters.huggingface.generate_imageleaderboard_scoresv1.0.0chp.adapters.huggingface.leaderboard_scoreslist_collectionsv1.0.0List HuggingFace Hub Collections — curated groupings of models and datasets. Enables agents to navigate thematic clusters (e.g.
chp.adapters.huggingface.list_collectionsload_datasetv1.0.0chp.adapters.huggingface.load_datasetmodel_cardv1.0.0chp.adapters.huggingface.model_cardpullv1.0.0Download any HuggingFace Hub artifact (model, dataset, tokenizer) to local cache via snapshot_download.
chp.adapters.huggingface.pullpull_for_local_llmv1.0.0chp.adapters.huggingface.pull_for_local_llmquantize_to_ggufv1.0.0chp.adapters.huggingface.quantize_to_ggufrun_pipelinev1.0.0chp.adapters.huggingface.run_pipelinesearch_datasetsv1.0.0Search HuggingFace Hub for datasets by task, sort, and filter. Symmetric to search_models — enables agents to discover datasets before loading them.
chp.adapters.huggingface.search_datasetssearch_modelsv1.0.0chp.adapters.huggingface.search_modelssearch_spacesv1.0.0Search HuggingFace Hub for Spaces (Gradio/Streamlit apps) by SDK, sort, and filter. Precondition for call_space discovery.
chp.adapters.huggingface.search_spacestokenizev1.0.0Encode text to token IDs or decode token IDs to text using a fast HuggingFace tokenizer. No full model load required.
chp.adapters.huggingface.tokenizetranscribe_audiov1.0.0chp.adapters.huggingface.transcribe_audio