generate-embeddings-uv-vllm.py

# /// script
# requires-python = ">=3.10"
# dependencies = [
#     "datasets",
#     "huggingface-hub[hf_transfer]",
#     "flashinfer-python",
#     "hf-xet>= 1.1.7",
#     "torch",
#     "transformers",
#     "vllm",
# ]
#
# ///
"""
Generate responses for prompts in a dataset using vLLM for efficient GPU inference.

This script loads a dataset from Hugging Face Hub containing chat-formatted messages,
applies the model's chat template, generates responses using vLLM, and saves the
results back to the Hub with a comprehensive dataset card.

Example usage:
    # Local execution with auto GPU detection
    uv run generate-responses.py \\
        username/input-dataset \\
        username/output-dataset \\
        --messages-column messages

    # With custom model and sampling parameters
    uv run generate-responses.py \\
        username/input-dataset \\
        username/output-dataset \\
        --model-id meta-llama/Llama-3.1-8B-Instruct \\
        --temperature 0.9 \\
        --top-p 0.95 \\
        --max-tokens 2048

    # HF Jobs execution (see script output for full command)
    hf jobs uv run --flavor a100x4 ...
"""

import argparse
import logging
import os
import sys
from datetime import datetime
from typing import Optional

from datasets import load_dataset
from huggingface_hub import get_token, login
from torch import cuda
from tqdm.auto import tqdm
from transformers import AutoTokenizer
from vllm import LLM
from dotenv import load_dotenv

# Enable HF Transfer for faster downloads
os.environ["HF_HUB_ENABLE_HF_TRANSFER"] = "1"

logging.basicConfig(
    level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s"
)
logger = logging.getLogger(__name__)


def check_gpu_availability() -> int:
    """Check if CUDA is available and return the number of GPUs."""
    if not cuda.is_available():
        logger.error("CUDA is not available. This script requires a GPU.")
        logger.error(
            "Please run on a machine with NVIDIA GPU or use HF Jobs with GPU flavor."
        )
        sys.exit(1)

    num_gpus = cuda.device_count()
    for i in range(num_gpus):
        gpu_name = cuda.get_device_name(i)
        gpu_memory = cuda.get_device_properties(i).total_memory / 1024**3
        logger.info(f"GPU {i}: {gpu_name} with {gpu_memory:.1f} GB memory")

    return num_gpus

def main(
    src_dataset_hub_id: str,
    output_dataset_hub_id: str,
    model_id: str = "Qwen/Qwen3-Embedding-0.6B",
    input_column: str = "text",
    output_column: str = "embeddings",
    gpu_memory_utilization: float = 0.90,
    input_truncation_len: Optional[int] = None,
    tensor_parallel_size: Optional[int] = None,
    max_samples: Optional[int] = None,
    hf_token: Optional[str] = None,
):
    """
    Main generation pipeline.

    Args:
        src_dataset_hub_id: Input dataset on Hugging Face Hub
        output_dataset_hub_id: Where to save results on Hugging Face Hub
        model_id: Hugging Face model ID for embedding generation
        input_column: Column name containing documents to embed
        output_column: Column name for generated embeddings
        gpu_memory_utilization: GPU memory utilization factor
        input_truncation_len: Maximum input length (None uses model default)
        tensor_parallel_size: Number of GPUs to use (auto-detect if None)
        max_samples: Maximum number of samples to process (None for all)
        hf_token: Hugging Face authentication token
    """
    generation_start_time = datetime.now().isoformat()

    # GPU check and configuration
    num_gpus = check_gpu_availability()
    if tensor_parallel_size is None:
        tensor_parallel_size = num_gpus
        logger.info(
            f"Auto-detected {num_gpus} GPU(s), using tensor_parallel_size={tensor_parallel_size}"
        )
    else:
        logger.info(f"Using specified tensor_parallel_size={tensor_parallel_size}")
        if tensor_parallel_size > num_gpus:
            logger.warning(
                f"Requested {tensor_parallel_size} GPUs but only {num_gpus} available"
            )

    # Authentication - try multiple methods
    load_dotenv()
    HF_TOKEN = hf_token or os.environ.get("HF_TOKEN") or get_token()

    if not HF_TOKEN:
        logger.error("No HuggingFace token found. Please provide token via:")
        logger.error("  1. --hf-token argument")
        logger.error("  2. HF_TOKEN environment variable")
        logger.error("  3. Run 'huggingface-cli login' or use login() in Python")
        sys.exit(1)

    logger.info("HuggingFace token found, authenticating...")
    login(token=HF_TOKEN)

    # Initialize vLLM
    logger.info(f"Loading model: {model_id}")
    vllm_kwargs = {
        "model": model_id,
        "tensor_parallel_size": tensor_parallel_size,
        "gpu_memory_utilization": gpu_memory_utilization,
        "task": "embed",
        "max_model_len": input_truncation_len + 128,
    }

    llm = LLM(**vllm_kwargs)

    # Load tokenizer for chat template
    logger.info("Loading tokenizer...")
    tokenizer = AutoTokenizer.from_pretrained(model_id)

    # Load dataset
    logger.info(f"Loading dataset: {src_dataset_hub_id}")
    dataset = load_dataset(src_dataset_hub_id, split="train")

    # Apply max_samples if specified
    if max_samples is not None and max_samples < len(dataset):
        logger.info(f"Limiting dataset to {max_samples} samples")
        dataset = dataset.select(range(max_samples))

    total_examples = len(dataset)
    logger.info(f"Dataset loaded with {total_examples:,} examples")

    # Determine which column to use and validate
    if input_column not in dataset.column_names:
        logger.error(
            f"Column '{input_column}' not found. Available columns: {dataset.column_names}"
        )
        sys.exit(1)
    logger.info(f"Using input column mode with column: '{input_column}'")

    # Process documents and truncate if specified
    logger.info("Preparing documents...")
    all_documents = []
    for example in tqdm(dataset, desc="Processing documents"):
        document = f"# {example['title_dl']}\n\nFrom: {example['source_url']}\n\n{example[input_column]}"
        # apply tokenizer to the document, then truncate using token counts
        if input_truncation_len is not None:
            tokens = tokenizer.encode(document)
            if len(tokens) > input_truncation_len:
                document = tokenizer.decode(tokens[:input_truncation_len])
        all_documents.append(document) # this is a list of strings

    # Generate embeddings - vLLM handles batching internally
    logger.info("vLLM will handle batching and scheduling automatically")
    outputs = llm.embed(all_documents)

    # Extract generated embeddings and create full response list
    logger.info("Extracting generated embeddings...")
    embeddings = [o.outputs.embedding for o in outputs]

    # Add responses to dataset
    logger.info("Adding responses to dataset...")
    dataset = dataset.add_column(output_column, embeddings)

    # Push dataset to hub
    logger.info(f"Pushing dataset to: {output_dataset_hub_id}")
    dataset.push_to_hub(output_dataset_hub_id, token=HF_TOKEN)

    logger.info("✅ Embedding generation complete!")
    logger.info(
        f"Dataset available at: https://huggingface.co/datasets/{output_dataset_hub_id}"
    )


if __name__ == "__main__":
    if len(sys.argv) > 1:
        parser = argparse.ArgumentParser(
            description="Generate responses for dataset prompts using vLLM",
            formatter_class=argparse.RawDescriptionHelpFormatter,
            epilog="""
Examples:
  # Basic usage with default Qwen model
  uv run generate-embeddings-uv-vllm.py input-dataset output-dataset
  
  # With custom model and parameters
  uv run generate-embeddings-uv-vllm.py input-dataset output-dataset \\
    --model-id Qwen/Qwen3-Embedding-0.6B \\
    --input-column text \\
    --output-column embeddings
  
  # Force specific GPU configuration
  uv run generate-embeddings-uv-vllm.py input-dataset output-dataset \\
    --tensor-parallel-size 2 \\
    --gpu-memory-utilization 0.95
  
  # Using environment variable for token
  HF_TOKEN=hf_xxx uv run generate-embeddings-uv-vllm.py input-dataset output-dataset
            """,
        )

        parser.add_argument(
            "src_dataset_hub_id",
            help="Input dataset on Hugging Face Hub (e.g., username/dataset-name)",
        )
        parser.add_argument(
            "output_dataset_hub_id", help="Output dataset name on Hugging Face Hub"
        )
        parser.add_argument(
            "--model-id",
            type=str,
            default="Qwen/Qwen3-Embedding-0.6B",
            help="Model to use for generation (default: Qwen3-Embedding-0.6B)",
        )
        parser.add_argument(
            "--input-column",
            type=str,
            default="text",
            help="Column containing text to embed (default: text)",
        )
        parser.add_argument(
            "--output-column",
            type=str,
            default="embeddings",
            help="Column name for generated embeddings (default: embeddings)",
        )
        parser.add_argument(
            "--max-samples",
            type=int,
            help="Maximum number of samples to process (default: all)",
        )
        parser.add_argument(
            "--input-truncation-len",
            type=int,
            help="Maximum input length (default: model's default)",
        )
        parser.add_argument(
            "--tensor-parallel-size",
            type=int,
            help="Number of GPUs to use (default: auto-detect)",
        )
        parser.add_argument(
            "--gpu-memory-utilization",
            type=float,
            default=0.90,
            help="GPU memory utilization factor (default: 0.90)",
        )
        parser.add_argument(
            "--hf-token",
            type=str,
            help="Hugging Face token (can also use HF_TOKEN env var)",
        )
        args = parser.parse_args()

        main(
            src_dataset_hub_id=args.src_dataset_hub_id,
            output_dataset_hub_id=args.output_dataset_hub_id,
            model_id=args.model_id,
            input_column=args.input_column,
            output_column=args.output_column,
            gpu_memory_utilization=args.gpu_memory_utilization,
            input_truncation_len=args.input_truncation_len,
            tensor_parallel_size=args.tensor_parallel_size,
            max_samples=args.max_samples,
            hf_token=args.hf_token,
        )
    else:
        # Show HF Jobs example when run without arguments
        print("""
vLLM Response Generation Script
==============================

This script requires arguments. For usage information:
    uv run generate-responses.py --help

Example HF Jobs command with multi-GPU:
    # If you're logged in with huggingface-cli, token will be auto-detected
    hf jobs uv run \\
        --flavor l4x4 \\
        https://huggingface.co/datasets/uv-scripts/vllm/raw/main/generate-responses.py \\
        username/input-dataset \\
        username/output-dataset \\
        --messages-column messages \\
        --model-id Qwen/Qwen3-30B-A3B-Instruct-2507 \\
        --temperature 0.7 \\
        --max-tokens 16384
        """)