Average Inference Latency Comparison in Milliseconds per Prediction

Average Inference Latency Comparison (milliseconds per prediction) is a dataset by Mohammed Saad Javeed. It compares inference latency across different models or systems, measured in milliseconds per prediction. The dataset was last updated on May 29, 2026, and is shared under a CC-BY-4.0 license.

Use Cases

• Compare inference speed across different machine learning models based on the latency metric.
• Benchmark hardware or software optimizations for model serving based on prediction latency.
• Analyze trade-offs between model accuracy and inference speed using the provided latency data.

Strengths

• Dataset is small (5.5 KB), facilitating quick download and inspection.
• Has a clear, permissive license (CC-BY-4.0) for reuse and sharing.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Mohammed Saad Javeed via figshare

Freshness

Last updated 2026-05-29 17:32:07; freshness should be verified.