🎯

huggingface

🎯Skill
g1joshi/agent-skills2 Installs9Last Updated Feb 10, 2026

Installation

$npx skills add g1joshi/agent-skills --skill huggingface

Run it in a terminal; it asks which AI coding tools to install it for (Claude Code, Codex, Cursor and more).

Uses skills.sh Β· Node.js 22.20+

After installing

Ask for related work and the agent uses it on its own, or call it yourself.

Run it (Claude Code)/huggingface
Install by hand

Copy the whole skill folder from GitHub to the place below. Copying only SKILL.md can leave out files the skill uses.

Open the skill folder β†’
  • This project .claude/skills/huggingface/
  • All projects ~/.claude/skills/huggingface/

These paths are for Claude Code. For other tools, the command above puts it in the right place.

Expert Hugging Face assistance covering Transformers, datasets, pipelines, Hub model loading, and tokenizers. Use when downloading, fine-tuning, and deploying open-source machine learning models.

Hugging Face

Hugging Face is the central platform and ecosystem for open-source machine learning, hosting models, datasets, Spaces, and foundational libraries like Transformers, Accelerate, and PEFT.

When to Use

  • Pretrained Model Discovery & Inference: Utilizing thousands of open-source vision, NLP, and audio models via transformers.
  • Parameter-Efficient Fine-Tuning (PEFT): Fine-tuning LLMs with LoRA / QLoRA using peft and bitsandbytes.
  • Dataset Management & Streaming: Accessing multi-gigabyte datasets without memory exhaustion via datasets.
  • Production Model Deployment: Deploying containerized endpoints with Hugging Face Text Generation Inference (TGI).

Quick Start

from transformers import pipeline

# One-line inference with modern pre-trained models
classifier = pipeline("sentiment-analysis", model="distilbert-base-uncased-finetuned-sst-2-english")

result = classifier("This framework makes NLP development seamless and fast!")
print(result) # [{'label': 'POSITIVE', 'score': 0.9998}]

Core Concepts

Inference Pipeline with Transformers & PyTorch

Loading state-of-the-art transformer models with automatic tokenization:

from transformers import pipeline
import torch

# Create accelerated text-classification pipeline
classifier = pipeline(
    task="text-classification",
    model="distilbert/distilbert-base-uncased-finetuned-sst-2-english",
    device=0 if torch.cuda.is_available() else -1
)

results = classifier([
    "This new AI development platform accelerates productivity significantly.",
    "The legacy system crashed repeatedly under moderate load."
])

for res in results:
    print(f"Label: {res['label']}, Score: {res['score']:.4f}")

4-Bit Model Quantization with BitsAndBytes

Loading large LLMs on consumer hardware using 4-bit NormalFloat (NF4):

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig

model_id = "meta-llama/Llama-3.2-3B-Instruct"

bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.bfloat16,
    bnb_4bit_use_double_quant=True
)

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    quantization_config=bnb_config,
    device_map="auto"
)

inputs = tokenizer("Explain quantum computing in one sentence:", return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens=50)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Streaming Massive Datasets with Datasets Library

Iterating over terabyte-scale datasets without downloading all files upfront:

from datasets import load_dataset

# Stream dataset lazily without full disk download
dataset = load_dataset('allenai/c4', 'en', split='train', streaming=True)

# Iterate through samples
for i, sample in enumerate(dataset.take(3)):
    print(f"Sample {i+1}: {sample['text'][:100]}...\n")

Common Patterns

Quantized Model Loading with BitsAndBytes (4-bit / 8-bit)

Problem: 7B-70B parameter models exceed single-GPU VRAM limits in FP16 precision.

Solution: Load with 4-bit NF4 quantization using BitsAndBytesConfig:

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig

bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.float16
)

model_id = "meta-llama/Llama-3.2-3B"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    quantization_config=bnb_config,
    device_map="auto"
)

Best Practices

Do:

  • Use BitsAndBytesConfig (nf4) or AWQ quantization to run open-weight LLMs with low memory consumption.
  • Set device_map="auto" when loading large models to distribute layers automatically across available GPUs and RAM.
  • Use datasets.load_dataset(..., streaming=True) for datasets that exceed local storage capacity.
  • Pass torch_dtype=torch.bfloat16 when running inference on modern Ampere/Hopper GPU architectures.

Don't:

  • Load entire models onto CPU before moving to GPU; instantiate directly with device_map.
  • Hardcode Hugging Face access tokens in code; authenticate via huggingface-cli login or HF_TOKEN.
  • Run unquantized 70B+ parameter models on consumer hardware; use quantized GGUF, EXL2, or 4-bit NF4.

Troubleshooting

ErrorCauseSolution
Gated repo error / 403 ForbiddenAccessing gated model (e.g. Llama/Mistral) without accepting license on Hub.Accept license on HuggingFace and run huggingface-cli login.
CUDA Out of Memory in from_pretrainedModel weights exceed GPU memory in FP32.Use device_map="auto", torch_dtype=torch.float16, or 4-bit quantization.
KeyError in tokenizer outputMismatched tokenizer and model configurations.Always load tokenizer and model from the identical model_id.

References

More from this repository10