Hugging Face

Fix Python type checking failures found locally or in CI, the automated checks run on code changes. It identifies affected files, applies targeted fixes to type declarations and checks, then reruns validation.

• Fix type checking failures reported by CI
• Resolve errors involving values that may be None
• Add a directory to automated type checking

Run diffusers generation pipelines from a terminal and save their outputs. Inspect accepted inputs without downloading model weights, apply LoRA adapters during generation, run remotely through HF Jobs, and inspect software and GPU environment details.

• Running one-off generation or batch jobs from a terminal
• Checking a new model's basic operation
• Inspecting accepted inputs before running a pipeline

Package an existing custom diffusers block into a directory ready for Hub upload. Select a class from a Python file and save its source, configuration, and associated files so others can load the block.

• Preparing a custom block for Hub distribution
• Selecting one block class from a Python file
• Creating the directory layout needed to load a shared block

Integrate a model or pipeline into diffusers, from file structure and weight conversion to tests and documentation. Verify numerical agreement with the original implementation, with modular pipelines as the preferred starting point for new integrations.

• Adding a new model and pipeline to diffusers
• Converting a pipeline to modular format
• Converting weights for a new version of a supported model

Review a diffusers contribution before opening a pull request. Check the full change against project rules, trace potentially unused code, and receive a severity-based report separating required fixes from points to discuss with reviewers.

• Checking a diffusers contribution before opening a pull request
• Investigating potentially unused code in a change
• Reviewing fixes again and preparing notes for reviewers

Check that changes to existing PEFT methods still work with saved training checkpoints. The skill guides compatibility tests, configuration checks, benchmark comparisons, and reporting, including comparisons of model outputs before and after a change.

• Checking compatibility after a forward method rewrite
• Changing state_dict structure while supporting old checkpoints
• Reviewing defaults for new configuration fields

Add or review research paper entries in TRL’s paper index. Organize each paper’s contribution, its connection to the implementation, and a Python configuration example using training settings supported by the paper.

• Documenting a newly implemented research method
• Checking paper index updates during PR review
• Recording training settings from a paper

Write and debug code for further training large language models with TRL. Match training methods to dataset formats, configure supervised or preference-based training, and set up reward functions through Python or the command line.

• Training a language model on example responses
• Training with preferred and rejected response pairs
• Debugging GRPO reward functions and batch settings

Train sentence-transformers models for search, similarity, and reranking, which reorders search results by relevance. Choose a model type, prepare data, configure training, compare results against a baseline, and publish the trained model to Hugging Face Hub.

• Fine-tuning an embedding model for domain-specific search
• Training a model to rerank search results
• Training SPLADE or ColBERT retrieval models

Build text, image, and audio features in JavaScript using machine learning models that run in browsers or server runtimes. Covers model selection, execution settings, caching, and memory cleanup without requiring a Python server.

• Add sentiment analysis to a JavaScript app
• Build image classification in a browser
• Implement audio transcription

Build and publish a browser demo for a supplied LoRA, a model adaptation for image or video tasks. Tailor the Gradio interface to its inputs and follow the model’s recommended settings on Hugging Face Spaces.

• Share a trained LoRA through a browser demo
• Build an interface tailored to image editing or video generation
• Apply recommended trigger words and generation settings

Find and run local AI models with llama.cpp. Choose GGUF model files and quantization levels that fit available memory, launch a command line session or local server, and test an OpenAI-compatible API on your computer.

• Find models to run on a local computer
• Choose quantization for available memory
• Launch and test a local OpenAI-compatible API

Check the AWS profile, region, account ID, and caller identity before starting AWS work. Read local configuration and validate credentials, then flag missing settings, expired access, or possible IAM restrictions associated with SSO authentication.

• Check the AWS account and region before deployment
• Identify the active profile when none was specified
• Detect expired credentials or missing configuration

Build and maintain machine learning apps on Hugging Face Spaces. Choose a suitable app framework and computing hardware, load models, deploy files, and check logs and outputs to diagnose build or runtime problems.

• Publishing a machine learning demo
• Adapting an existing Gradio app for ZeroGPU
• Diagnosing Space build or runtime failures

Create interactive web demos and chat interfaces in Python with Gradio. Connect functions to text, image, audio, and file controls, arrange screen layouts, and define what happens when users click buttons or change inputs.

• Turning a Python function into an interactive demo
• Building a chatbot interface
• Adding image and audio inputs and outputs

Prepare an isolated Python environment for SageMaker deployments and AWS automation. The skill selects a supported Python version, installs current AWS packages, and checks versions so installation and execution use the same environment.

• Prepare a Python environment for SageMaker deployment
• Separate AWS automation dependencies from system Python
• Check installed AWS package versions

Deploy SageMaker endpoints, the services that answer model requests, with automatic capacity scaling, CloudWatch alarms, and resource tags. Supports real-time and asynchronous inference, scaling down to zero instances, deployment checks, and resource cleanup.

• Deploying a real-time model endpoint
• Scaling a sparse workload down to zero instances
• Running long inference jobs asynchronously

Explore Hugging Face datasets through the Dataset Viewer API. Preview and retrieve rows, search text, filter records, inspect sizes and column statistics, and find Parquet download links for further analysis or data extraction.

• Preview dataset contents and data splits
• Search text and filter dataset rows
• Inspect dataset sizes, statistics, and Parquet links

Read and summarize AI research papers from Hugging Face or arXiv links and identifiers. Retrieve paper content alongside structured details about authors, related models, datasets, Spaces, and implementation links to support further investigation.

• Summarizing a paper from its URL or arXiv ID
• Finding models, datasets, and Spaces linked to a paper
• Checking authors and implementation links

Train or fine-tune language and vision models on Hugging Face Jobs without a local GPU. Covers dataset checks, hardware and cost planning, training monitoring, Hub storage, and GGUF conversion for local use.

• Fine-tune a language model without a local GPU
• Validate preference data before DPO training
• Estimate training costs and choose GPU hardware

Manage Hugging Face Hub through the hf command line tool. Download and upload models or datasets, organize cloud storage, manage authentication and local caches, run scheduled jobs, and deploy or maintain model inference services.

• Download or publish models and datasets
• Sync training checkpoints to cloud storage
• Inspect and clean the local cache

Find AI models for a specific task using official Hugging Face benchmark leaderboards. Compare scores, model sizes, and licenses, with memory-based estimates of device suitability and flags for models requiring quantization or API access.

• Compare coding or reasoning models by benchmark scores
• Find model candidates for a laptop's available memory
• Choose models for OCR or speech recognition

Write and debug code for further training large language models with TRL. Match training methods to dataset formats, configure supervised or preference-based training, and set up reward functions through Python or the command line.

• Training a language model on example responses
• Training with preferred and rejected response pairs
• Debugging GRPO reward functions and batch settings

Train sentence-transformers models for search, similarity, and reranking, which reorders search results by relevance. Choose a model type, prepare data, configure training, compare results against a baseline, and publish the trained model to Hugging Face Hub.

• Fine-tuning an embedding model for domain-specific search
• Training a model to rerank search results
• Training SPLADE or ColBERT retrieval models