This page shows the empirical validation behind SOPH published draw proofs and other audit-heavy workflows. Every chart and log below is generated from actual API output, not synthetic demos or marketing screenshots.
Live battery runs and analysis views update continuously. This is the empirical validation behind SOPH published draw proofs and audit-heavy workflows.
Six independent test batteries running continuously. Every byte tested before it reaches a customer.
Each cell is one named test. Colour shows the weighted pass/warn/fail ratio. Hover for details.
Deeper batch-level metrics for teams that want more than pass/fail battery summaries.
Metrics from the latest test run.
Measures randomness and unpredictability of generated output
Percentage of output bits that flip when changing 1 input bit
Percentage of output bits that flip when changing 1 key bit
Statistical spread of byte distribution
Percentage of 1s in the bitstream (should be ~50%)
Average length of consecutive identical bits (ideal: 2.0 for random; status from z-score)
Join the waitlist for launch access to verifiable draws. If launch-offer capacity remains, we'll confirm it by email.