[RadNi]

Integrate committed setup recursion and the complete portable Akita verifier optimization stack: compact deferred setup, audited selected-row catalog packets, serial guest challenge decoding, byte-string advice transport, the size-class allocator, and optional transcript-authenticated quartet and position hints. Include advice preparation tooling and pin the matching Akita integration.

The original complete stack measured 397,276,220 guest cycles with field and NTT inlines disabled, with ten committed proofs generated and verified. This publication port has not been rebenchmarked.

Validation: formatting, metadata and static checks passed. Rust tests and Clippy remain pending.

[github-actions]

📏 PR diff

Note

Actual code changed: 2,656 lines

🟢 +2,568 added    🔴 −88 removed

Tests, docs, fixtures, and helper scripts excluded.

Category	🟢 Added	🔴 Removed
⌨️ Actual code	+2,568	−88
🧪 Tests	+363	−48
📚 Docs	+283	0
📦 Fixtures	+109	−16
🔧 Helper scripts	+4,954	−4,768
Total diff	+8,277	−4,920

_{Every changed line is classified once.}

[github-actions]

Benchmark comparison (crates)

Same-runner A/B: base_run = 54f30ce (merge target) vs pr_run = 71233cc (PR head merged onto it). Bold = faster run; Δ colored when the difference exceeds the ±10% noise threshold.

No differences above the ±10% noise threshold.

All 92 benchmarks

Benchmark	base_run (merge target)	pr_run (this PR)	Δ
EqPolynomial::evals/17	1.6±0.03ms	1.6±0.06ms	+2.1%
EqPolynomial::evals/19	6.1±0.09ms	6.0±0.12ms	-1.5%
EqPolynomial::evals/20	15.2±0.27ms	15.4±0.34ms	+1.1%
EqPolynomial::evals/22	60.6±1.15ms	60.0±1.12ms	-1.1%
EqPolynomial::evaluations/14	435.1±8.47µs	440.4±10.26µs	+1.2%
EqPolynomial::evaluations/18	5.4±0.07ms	5.4±0.06ms	+0.6%
EqPolynomial::evaluations/20	29.3±1.21ms	29.3±1.02ms	+0.0%
F128/accumulator/deferred	1.3±0.03µs	1.3±0.02µs	+0.2%
F128/accumulator/naive	2.6±0.06µs	2.6±0.06µs	-0.1%
F128/mul	2.1±0.03ns	2.1±0.04ns	-0.1%
F128/mul_slice	2.1±0.04µs	2.1±0.05µs	+0.2%
F128/square	2.8±0.05ns	2.8±0.06ns	+0.9%
F192/accumulator/deferred	2.8±0.07µs	2.8±0.05µs	-0.3%
F192/accumulator/naive	4.0±0.08µs	4.0±0.07µs	-0.1%
F192/mul	3.6±0.05ns	3.6±0.07ns	+0.9%
F192/mul_slice	3.7±0.06µs	3.7±0.06µs	+0.9%
F192/square	2.7±0.05ns	2.7±0.06ns	+0.9%
F64/accumulator/deferred	702.9±12.84ns	703.4±11.91ns	+0.1%
F64/accumulator/naive	926.3±23.72ns	923.6±17.80ns	-0.3%
F64/mul	0.9±0.02ns	0.9±0.02ns	+0.7%
F64/mul_slice	920.3±14.84ns	924.2±13.29ns	+0.4%
F64/square	0.9±0.02ns	0.9±0.01ns	+0.7%
Polynomial::bind/14	113.3±4.62µs	114.5±7.98µs	+1.0%
Polynomial::bind/18	1.7±0.02ms	1.7±0.03ms	+0.8%
Polynomial::bind/20	6.7±0.12ms	6.7±0.19ms	-0.5%
Polynomial::bind_low_to_high_in_place/Fr/14	174.9±4.35µs	172.0±5.31µs	-1.7%
Polynomial::bind_low_to_high_in_place/Fr/20	6.9±0.16ms	6.9±0.16ms	-0.1%
Polynomial::bind_low_to_high_in_place/Fr/24	113.7±0.86ms	113.3±1.94ms	-0.4%
Polynomial::bind_low_to_high_in_place/Prime128OffsetA7F7/14	47.0±0.90µs	46.5±0.90µs	-0.9%
Polynomial::bind_low_to_high_in_place/Prime128OffsetA7F7/20	1.7±0.04ms	1.7±0.10ms	+1.3%
Polynomial::bind_low_to_high_in_place/Prime128OffsetA7F7/24	24.6±0.42ms	24.6±0.75ms	+0.0%
Polynomial::evaluate/20	39.1±0.85ms	39.6±1.18ms	+1.2%
UnivariatePoly::from_evals/16	41.9±0.76µs	41.9±0.74µs	-0.2%
UnivariatePoly::from_evals/3	5.2±0.12µs	5.2±0.09µs	-0.4%
UnivariatePoly::from_evals/4	7.8±0.13µs	7.8±0.14µs	+0.8%
UnivariatePoly::from_evals/8	18.9±0.37µs	19.0±0.39µs	+0.4%
append_bytes/Blake2b/256B	333.9±9.31ns	328.3±9.33ns	-1.7%
append_bytes/Blake2b/32B	74.5±4.31ns	74.8±3.11ns	+0.4%
append_bytes/Keccak/256B	675.0±15.99ns	686.8±24.46ns	+1.7%
append_bytes/Keccak/32B	63.8±3.39ns	64.3±3.64ns	+0.8%
append_bytes/Poseidon/256B	186.2±3.44µs	184.3±4.51µs	-1.0%
append_bytes/Poseidon/32B	57.1±1.37µs	56.7±1.35µs	-0.7%
challenge/Blake2b	751.5±20.57ns	745.8±20.89ns	-0.8%
challenge/Keccak	473.5±14.83ns	470.4±11.15ns	-0.7%
field arithmetic/add/fp128_a7f7	4.3±0.09µs	4.3±0.07µs	-0.5%
field arithmetic/add/fp32_c99	2.5±0.06µs	2.5±0.05µs	-0.3%
field arithmetic/add/fp64_c59	2.5±0.06µs	2.5±0.06µs	-0.7%
field arithmetic/mul/fp128_a7f7	11.8±0.19µs	11.7±0.19µs	-0.7%
field arithmetic/mul/fp32_c99	4.4±0.11µs	4.4±0.08µs	+0.6%
field arithmetic/mul/fp64_c59	5.3±0.09µs	5.4±0.14µs	+1.1%
field arithmetic/square/fp128_a7f7	14.0±0.22µs	14.2±0.34µs	+1.5%
field arithmetic/square/fp32_c99	4.6±0.09µs	4.7±0.13µs	+0.4%
field arithmetic/square/fp64_c59	4.7±0.08µs	4.7±0.10µs	-0.1%
field arithmetic/sub/fp128_a7f7	3.3±0.06µs	3.4±0.08µs	+0.4%
field arithmetic/sub/fp32_c99	1.4±0.03µs	1.4±0.03µs	+0.2%
field arithmetic/sub/fp64_c59	1.8±0.05µs	1.8±0.04µs	-1.7%
field reduction/reduce_product/fp128_a7f7	19.4±0.37µs	19.4±0.55µs	-0.1%
fp64_prime63_offset259/ext2_packed_mul	20.1±0.42ns	20.1±0.52ns	-0.1%
fp64_prime63_offset259/ext2_scalar_mul	4.5±0.20ns	4.5±0.06ns	-0.2%
fp64_prime63_offset259/packed_mul	7.4±0.16ns	7.3±0.17ns	-0.7%
fp64_prime63_offset259/scalar_mul	1.6±0.03ns	1.5±0.04ns	-0.6%
fp64_prime64_offset59/ext2_dependent_mul_1024	7.8±0.14µs	7.8±0.14µs	-0.4%
fp64_prime64_offset59/ext2_neg_one_scalar_mul	5.3±0.14ns	5.2±0.13ns	-1.1%
fp64_prime64_offset59/ext2_packed_mul	25.6±0.54ns	25.2±0.36ns	-1.7%
fp64_prime64_offset59/ext2_scalar_mul	5.5±0.11ns	5.4±0.14ns	-1.4%
fp64_prime64_offset59/ext2_scalar_square	4.6±0.10ns	4.6±0.13ns	-1.1%
g1_add	314.0±5.00ns	314.5±6.33ns	+0.2%
g1_deserialize_bincode	6.9±0.15µs	6.9±0.14µs	+0.3%
g1_double	155.3±3.23ns	155.5±2.35ns	+0.2%
g1_msm/1024	10.5±0.20ms	10.3±0.13ms	-1.7%
g1_msm/16	447.3±11.20µs	444.1±11.28µs	-0.7%
g1_msm/256	3.5±0.06ms	3.5±0.05ms	-0.5%
g1_msm/4	202.1±7.30µs	197.2±8.14µs	-2.4%
g1_msm/64	1.3±0.03ms	1.3±0.03ms	-1.3%
g1_scalar_mul	49.5±1.29µs	49.1±0.86µs	-0.9%
g1_serialize_bincode	113.4±2.42ns	113.3±1.82ns	-0.2%
g2_msm/16	1.6±0.03ms	1.7±0.03ms	+0.8%
g2_msm/256	13.3±0.19ms	13.1±0.17ms	-1.7%
g2_msm/4	673.4±13.53µs	666.0±11.52µs	-1.1%
g2_msm/64	4.8±0.06ms	4.7±0.08ms	-1.2%
g2_scalar_mul	277.1±7.61µs	272.8±6.65µs	-1.6%
gt_scalar_mul	808.1±23.79µs	796.6±13.59µs	-1.4%
multi_pairing/16	5.1±0.06ms	5.1±0.06ms	+0.1%
multi_pairing/2	1.1±0.03ms	1.1±0.03ms	+0.1%
multi_pairing/4	1.7±0.03ms	1.7±0.04ms	+0.3%
multi_pairing/8	2.8±0.04ms	2.8±0.04ms	+0.3%
pairing	849.3±16.91µs	854.2±18.91µs	+0.6%
pedersen_commit/1024	10.5±0.11ms	10.3±0.14ms	-1.8%
pedersen_commit/16	508.9±24.05µs	504.2±7.74µs	-0.9%
pedersen_commit/256	3.5±0.05ms	3.5±0.05ms	+0.6%
pedersen_commit/4	249.7±5.38µs	244.8±4.54µs	-2.0%
pedersen_commit/64	1.4±0.02ms	1.4±0.03ms	-0.1%