Skip to content
Other

yolo26s-construction-ppe

construction site personal protective equipment detector

2 min read

This model is designed to automatically detect people, safety helmets, high-visibility vests, and safety violations (such as missing helmets or vests) in construction site photos. It is built on the YOLO architecture (a popular real-time computer vision library) and is available in PyTorch and ONNX formats with various optimization levels (FP32, FP16, INT8). This development will be useful for occupational health and safety engineers, video monitoring system developers, and automation specialists for construction site monitoring.

Key features

  • Accurately detects helmets, vests, and people in images.
  • Supports fast ONNX FP16 inference with response times down to 6.3 ms on an NVIDIA Tesla T4 GPU.
  • An INT8 quantized version weighing only 11 MB is available for running on CPUs and edge devices.
  • Rigorous dataset verification was performed to prevent data leakage between training and test splits.

What you can do

  • Integrate the model into a construction site video surveillance system to automatically detect workers without helmets.
  • Set up a photo analysis script for safety inspectors' mobile devices.
  • Run the optimized ONNX version on resource-constrained edge boxes with limited CPU resources.

Example

from ultralytics import YOLO

model = YOLO("best-int8.onnx")
for r in model.predict("site.jpg", conf=0.35):
    for xyxy, cls, conf in zip(r.boxes.xyxy.tolist(), r.boxes.cls.tolist(), r.boxes.conf.tolist()):
        print(r.names[int(cls)], round(conf, 2), [round(v) for v in xyxy])
Read moreDetails
Metrics on the test split (119 images): overall mAP50 is 0.900, mAP50-95 is 0.509.
Hardware requirements: GPU execution is supported (FP16 recommended) along with CPU via ONNX Runtime.
Limitations: rare classes like no-helmet have few examples in the test set, requiring validation on live security cameras.
License: weights inherit from Ultralytics YOLO and require compliance with AGPL-3.0 for commercial use.
Examples
Discuss Telegram 1

Will it run?

Runs in FP16

Fits without quantisation.

  • FP16weights 0.0 GB · ready files0.5 GB
  • FP8 / INT8recommendedweights 0.0 GB0.5 GB
  • 4-bitweights 0.0 GB0.5 GB

Tip: go 8-bit.FP8 / INT8 is within about a percent of full precision in quality, takes half the memory and runs faster on RTX 40/50 and H100. Full precision only pays off for fine-tuning.

What to get for this model

Massed ComputeOpen Massed Compute

Your own GPU: FP8 / INT8 — from RTX 2060 · FP16 — from RTX 2060

Massed Compute hourly prices, November 2026. The discounted price applies with the coupon at checkout.

Estimate: weights plus working memory. Long videos, high resolutions and long contexts need more.

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # mpa-17/yolo26s-construction-ppe
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download mpa-17/yolo26s-construction-ppe --local-dir ~/models/yolo26s-construction-ppe
    # The model is in ~/models. How to run it is in the model card (link above).
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

New posts about versions of yolo26s-construction-ppe (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

mpa-17/yolo26s-construction-ppe
Author
mpa-17
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

01

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments