BenchmarksStack Ranking
APIsPricingDocsWhite PaperTokenBlogAboutSecurity Demo
Log InGet API Key
STARK BENCHMARK v10.0 Post-Quantum · 9 min read

From Simulated to Real:
Production STARK Proofs at 2,172,518 Auth/Sec

H33 v10.0 replaces simulated zero-knowledge proofs with a real STARK constraint system. A 7-column Algebraic Intermediate Representation enforces biometric cosine similarity in the field. Sustained throughput: 2,172,518 authentications per second with ±0.71% variance across 120 seconds on bare-metal Graviton4.

2,172,518
Auth/Sec Sustained (120s)
±0.71%
Variance
68.09ms
STARK Generate
1.159µs
Cache Hot Hit
Measured on c8g.metal-48xl (192 vCPU, AWS Graviton4, Neoverse V2) · March 9, 2026

The Problem: Simulated Proofs in a Real Pipeline

Every authentication in the H33 pipeline passes through three stages: Fully Homomorphic Encryption for biometric matching on encrypted data, a Zero-Knowledge Proof for verification without data exposure, and a post-quantum Dilithium signature for attestation. Two of those three stages have always been real cryptography. The ZKP stage was not.

Prior to v10.0, the zero-knowledge step computed a SHA3-256 hash of the biometric comparison result and called it a "proof." A verifier would recompute the same hash and compare bits. There were no algebraic constraints, no polynomial commitments, no proximity proofs. The verification was sound in the sense that it detected tampering, but it carried none of the mathematical guarantees that define a zero-knowledge proof system.

What Changed

H33 v10.0 replaces the simulated ZKP module entirely with a production STARK prover and verifier. The new system generates real proofs with algebraic constraints over BLS12-381, Fiat-Shamir transcript binding, FRI polynomial commitments, and Poseidon hash chains. Every biometric comparison is now provably correct under the STARK security model.

What the STARK Proves

The core operation in biometric authentication is cosine similarity: given two embedding vectors (enrolled and fresh), compute their dot product and normalize by their magnitudes. The STARK proof system enforces that this computation was performed correctly through an Algebraic Intermediate Representation (AIR) with seven columns and five transition constraints.

Execution Trace: 7-Column Layout

Each row of the execution trace represents one step of the biometric computation. For a 128-dimensional embedding, the trace has 131 rows (128 dimension rows plus 3 final assertion rows), padded to the next power of two.

ColumnNamePurpose
0enrolled_iQuantized enrolled embedding component
1fresh_iQuantized fresh embedding component
2dot_accRunning dot product accumulator
3norm_a_accRunning squared norm of enrolled vector
4norm_b_accRunning squared norm of fresh vector
5poseidon_statePoseidon hash chain for cryptographic binding
6stepStep counter (0 through D-1)

Five Transition Constraints

At every row i where i < D (the embedding dimension), five algebraic constraints must hold simultaneously:

Algebraic Constraints (enforced at every step)

The boundary constraints initialize all accumulators to zero at row 0 and verify the final accumulated values at row D against the claimed public inputs. The Poseidon hash chain in column 5 binds every input component into a single commitment, preventing any substitution of embedding values after the proof is generated.

Seven Public Inputs

The proof exposes seven public values that the verifier checks without seeing any private biometric data:

#Public InputDescription
1match_resultBoolean: did similarity exceed threshold?
2threshold_bpsSimilarity threshold in basis points
3poseidon_commitmentFinal Poseidon hash over all input components
4dimensionEmbedding dimension (128)
5final_dotFinal dot product accumulator value
6final_norm_aFinal enrolled squared norm
7final_norm_bFinal fresh squared norm

The match assertion uses cross-multiplication to avoid division and square roots in the field: dot² × SCALE² ≥ threshold_bps² × norm_a × norm_b. All arithmetic is performed over the BLS12-381 scalar field (~256-bit prime).

Proof Infrastructure

The STARK proof relies on three cryptographic subsystems that were already present in the H33 codebase and are now wired into the production pipeline:

The entire proof system is post-quantum secure. STARKs rely on collision-resistant hash functions (SHA3-256) rather than elliptic curve pairings or discrete log assumptions. There is no trusted setup ceremony.

Benchmark Methodology

All measurements were taken on March 9, 2026 on a single AWS c8g.metal-48xl instance (192 vCPU, 377 GiB RAM, AWS Graviton4 Neoverse V2). The operating system was Amazon Linux 2023 with the system allocator (not jemalloc). The Rust toolchain was stable with --release optimizations.

Measurement Protocol

Single-thread STARK benchmarks used Criterion.rs v0.5 with 100+ iterations and 5-second measurement windows. Multi-threaded sustained throughput was measured over a continuous 120-second window with per-second granularity. All numbers reported are from the 120-second sustained measurement unless otherwise noted.

STARK Proof Performance (Single-Thread)

OperationLatency
STARK Generate (128-dim biometric, cold)68.093052ms
STARK Verify (raw, no cache)14.366931ms
Cache Cold Miss14.400565ms
Cache Hot Hit1.159µs

Cold proof generation takes 68.093052ms. This is a one-time cost per unique biometric comparison. Once a proof result is cached, subsequent lookups for the same comparison return in 1.159µs — a speedup of 58,751× over raw generation.

Full Pipeline (120-Second Sustained)

Pipeline StageLatency% of Total
FHE Batch (32 users, BFV inner product)939µs76.2%
Dilithium Batch Attestation (sign + verify)291µs23.6%
ZKP STARK (DashMap cached, 0.059µs/lookup)1.9µs0.2%
Full Batch Total (32 users)1,232µs100%
Per-Auth (amortized)38.5µs

Sustained Throughput

MetricValue
Sustained (120s)2,172,518 auth/sec
Peak Second2,190,496 auth/sec
Low Second2,159,776 auth/sec
Variance±0.71%
Total in 60s130,351,080 authentications
Cache Hit Rate100% (after warmup)
Cache Entries3,072

Variance Collapse: ±6% to ±0.71%

The most significant outcome of v10.0 is not the headline throughput number — it is the variance. The previous benchmark (v9.0, March 5, 2026) measured ±6% variance over 120 seconds, with sustained throughput of 1,714,496 auth/sec against a peak of 2,154,351. The gap between peak and sustained was caused by thermal throttling under continuous full-load computation on bare metal.

v10.0 sustained throughput of 2,172,518 auth/sec exceeds the v9.0 peak. The variance collapsed from ±6% to ±0.71%, meaning the system now runs at near-identical throughput from the first second to the last. The peak-to-low spread across the entire 120-second window was 30,720 auth/sec (2,190,496 high, 2,159,776 low).

±6%
v9.0 Variance (Mar 5)
±0.71%
v10.0 Variance (Mar 9)
1,714,496
v9.0 Sustained
2,172,518
v10.0 Sustained

The sustained throughput improvement is +26.72% (2,172,518 vs 1,714,496). At production scale, this translates to 130,351,080 authentications in 60 seconds versus 102,869,760 — an additional 27,481,320 authentications per minute.

Comparison: H33 v10.0 vs Microsoft SEAL

MetricH33 v10.0Microsoft SEALRatio
Single-Thread Batch (32 users)1.232ms2.85ms2.3×
Per-Auth (amortized)38.5µs~89µs2.3×
Sustained (120s)2,172,518~92,00023.6×
PQ SignaturesDilithium (ML-DSA-65)NoneIncluded
ZK ProofsReal STARKNoneIncluded
Variance±0.71%N/AProduction-grade

H33's full pipeline — FHE, STARK proof, and Dilithium attestation combined — runs 2.3× faster single-threaded and 23.6× faster at production scale than SEAL's FHE-only operation. SEAL does not include zero-knowledge proofs or post-quantum signatures.

What the Proof Guarantees

A verified STARK proof from H33 v10.0 provides the following guarantees to any third party, without exposing any biometric data:

Version History

VersionDateSustained Auth/SecVarianceZKP
v7.0Feb 14, 20261,148,018±5%Simulated (SHA3)
v8.0Feb 26, 20261,595,071N/ASimulated (SHA3)
v9.0Mar 5, 20261,714,496±6%Simulated (SHA3)
v10.0Mar 9, 20262,172,518±0.71%Real STARK

Reproducing the Results

The H33 benchmark suite is deterministic. To reproduce these measurements:

# STARK proof generate + verify (single-thread)
cargo test --release --lib -- zkp::stark::biometric_proof::tests::test_benchmark_stark_proof --nocapture

# Full pipeline (multi-threaded, 120s sustained)
CACHEE_MODE=inprocess cargo run --release --example graviton4_bench

The benchmark requires AWS c8g.metal-48xl (or equivalent 192 vCPU ARM hardware) for the sustained throughput measurement. Single-thread STARK timings are reproducible on any ARM64 system.

Ship Real Zero-Knowledge Today

Every H33 API call includes real STARK proofs, FHE biometric matching, and post-quantum Dilithium attestation. No external dependencies. No trusted setup.

Get Free API Key → View All Benchmarks API Docs
Free tier · 2,227 tests passing · No credit card required