AWS Neuron docs: YOLOv4 on inf1.2xlarge real-time, 180 img/s (p50 40ms)
180.0 images/s (batch 1, fp32)
Fields
id
benchmarks/aws-neuron-inf1-yolov4-rt-inf-sMetric
tps_aggregate
Value
180
Unit
images/s (batch 1, fp32)
Model
YOLOv4
Methodology
VENDOR-MEASURED (AWS Neuron docs, Inf1 Inference Performance, NeuronPerf, last update 2024-09-16, Neuron 2.20). inf1.xlarge = 1 Inferentia1 chip (4 NeuronCores). Encoder/CNN models only — NO generative LLM rows exist for Inf1. Units are inferences/sec (BERT, seq_len 128) or images/sec (ResNet/YOLO), NOT tokens/sec. PyTorch 1.13, inf1.2xlarge, batch 1, Real-Time run-mode, fp32.
Measured by
vendor
Reproducible
yes
notes
Latency-optimized vision row. No LLM/generative data exists for Inferentia1 — the product line's LLM story starts at Inferentia2.
Notes
Latency-optimized vision row. No LLM/generative data exists for Inferentia1 — the product line's LLM story starts at Inferentia2.