FocsleFo’c’sle
HA

Hailo-YOLO27-INT8

11%
by Hailo

Vendor-tuned YOLO27 for the Hailo Dataflow Compiler. Ships as native HEF.

Object detectionApache-2.0INT8hailovendor-tunedcoco
78K downloads 12K deploymentsUpdated Apr 8, 2028
Headline:11.4ms · Hailo-8 · INT8
OverviewBenchmarks3Sim ResultsDeploy3FilesDiscussion23

About this model

Vendor-tuned YOLO27 for the Hailo Dataflow Compiler. Ships as native HEF.

Authored by hailo. Curated into the Fo’c’sle reference set on 2028-04-08. All cross-chip benchmarks below were collected in matched-pair runs in the HIL lab using the same input pipeline, same upstream preprocessing, and the same downstream consumer. See the methodology page for the full protocol.

Task
Object detection
Parameters
23.4 M
Benchmarked on
3 chips
Deployments
12K

Architecture

Detection backbone + neck + head
Inferred from upstream weights · simplified
ImageCSPDarknet53PANet neckCls headBox headObj headNMS · DFL

Headline benchmarks

HHailo-10HINT8
16.0ms p50
62 FPS97.9% acc3.3 W
πRaspberry Pi 5 + Hailo HATINT8
22.3ms p50
45 FPS98.9% acc7.1 W
HHailo-8INT8
22.7ms p50
44 FPS98.7% acc2.2 W

Training data

Pretrained on the upstream maintainer’s released checkpoint. Edge-distillation pass uses 2.4M frames from the Fo’c’sle distillation corpus (consented public data + opt-in publisher contributions). Quantization-aware fine-tune uses 320K calibration samples drawn from the target task’s eval domain.

  • Pretraining corpus: upstream maintainer release
  • Distillation corpus: 2,400,000 frames
  • Calibration set: 320,000 samples (per task)
  • Eval set: standard benchmark + matched-pair HIL runs