einsum-opt_size on zen5-cpu1620fa86da59cefb652053e2b9012ec08dbfd5a0defaultcpueinsum (the provider in the measured revision; see library above)1f7ba58132b0086bed6ebbc98923990fdb93ac8f0d31220ef234a3c8eaebd79b1f28eb8e3675be786599cc4030810368a6b7964b054b9e9f9242fae601d2e87abf1635d1f5940523zen5-cpudevelopment — provisional: recorded on a development host, superseded by the measurement host's dataset2026-10-20T20:58:47.584608Zscripts/v2/record_run.py zen5-cpu einsum-opt_sizedata/v2-results/cpueinsum/zen5-cpu/einsum-opt_size/20261020T205611Z/tenferro-einsum-benchmark problem set: pairwise matrix products, batched and elementwise products, outer products, a matrix-chain problem, MPS inner products, two MERA networks, two tensor-network permutation problems, and two likelihood contractions of 38 and 84 operands. The operands are generated by the harness (deterministic, dtype-templated) and the contraction order is taken from the corpus's own paths entry, so the rows measure one library against two declared schedules, not two libraries.
Source: The corpus is tenferro-einsum-benchmark data/instances/*.json at the revision in pins/tenferro-einsum-benchmark.rev, materialised by scripts/setup_extern_deps.sh. Each instance declares name, format_string, shapes, dtype and two opt_einsum-style positional paths; crates/cpueinsum-bench/src/corpus.rs in the measured revision translates them into cpueinsum's integer-label, SSA-order API.
Known blind spot: The corpus instance str_mps_varying_inner_product_200 is not declared: the harness's own cross-row verification refuses it (MISMATCH(opt_flops) rel_err=NaN, exit 1) at every declared thread budget. It is therefore not measured, not skipped, and not part of this suite's coverage; the evidence is in docs/verify-refusal-str_mps-varying-inner-product-200.md.
Known blind spot: The declared identity is the numeric sentinel n = 1: every instance has fixed extents, so the declaration grid has one shape and the real operand shapes are the harness's case_m (operand count), case_n (total input elements) and case_k (output elements) columns, not geometry the toolkit compares.
Known blind spot: The engine rows are two contraction orders of the same library, so a row here is a schedule comparison inside cpueinsum, not a comparison against an independent implementation; there is no reference engine and no vendor arm in the first record.
Known blind spot: Every instance has unit batch, so a cost proportional to batch elements is invisible.
Known blind spot: One size per instance: a ratio here says nothing about a neighbouring instance, and the set is not a sweep.
Known blind spot: A corpus revision is not one of the measured repository's invalidated_by paths: the declaration records the corpus pin it was generated from, and a corpus bump has to be a deliberate regeneration of this suite.opt_size — cpueinsum executing the corpus's paths.opt_size order: the order the corpus declares as minimizing the largest intermediate (memory).
Identity: the measured cpueinsum-rs revision in this run's manifest (library.commit), pinned by pins/cpueinsum-rs.rev
Caveat: The order is the caller's, taken verbatim from the corpus; no order is searched or corrected here.
As measured: cpueinsum (the provider in the measured revision; see library above)AMD Ryzen AI 9 HX 4702424 MiB (2 instances)0-3: 16 MiB shared; 4-11: 8 MiB sharedLinux 7.0.0-38-generic / x86_64shinaoka-EVO-X1Pro4, 4T -> 4-7Every timed cell matches a successful verification census, including operation variants.
Each cell is median of process medians ± sample SD of those medians; W is the median
within-process sample SD. All values are milliseconds. These are descriptive spreads,
not confidence intervals. The between-process spread includes the rotating-order schedule.
Each process retains 5 chronological calls; 2 complete process sets are recorded.
Spread guide (computed from this run's own rows): per-engine between-start range median 0.48%, worst 6.65% among cells of 1 us or more (a sub-microsecond row is timer and scheduler jitter, not kernel time); 1 of 26 engine cells exceed 5% and carry !; median within-process relative SD 0.50%. A cell whose ±, ! or p is as large as the difference you are reading is not evidence of a regression.
no row, so no ratio there is position-balanced). ! marks a cell where some engine's own
per-start median range exceeds 5% of its median, and p a balanced cell whose estimated
position-effect ratio exceeds 1.05. Flagged cells are retained, not discarded; the
between-start variation can exceed the difference being compared.
! between-start range over 5%, p position effect over 5%, dispatch-dominated means the
row is under 20 us so 4T/8T pay thread-region start-up). Per-start ratios and every
chronological sample are in the run directory named at the top of this page (run*.csv).
These describe the recorded revisions; they are not a regression claim.
| case | opt_size (ms) | notes |
|---|---|
| bin_batched_matmul_b32_m64_n64_k64/n1/b1 | 0.555303 ± 0.00224577 (W 0.00203681) | |
| bin_elementwise_mul_2048x2048/n1/b1 | 4.45272 ± 0.0123425 (W 0.0107548) | |
| bin_matmul_256/n1/b1 | 1.07435 ± 0.00548291 (W 0.000420243) | |
| bin_outer_product_4096/n1/b1 | 6.1783 ± 0.0427524 (W 0.0207439) | |
| gm_queen5_5_3.wcsp/n1/b1 | 1640.93 ± 1.317 (W 1.65217) | |
| lm_batch_likelihood_brackets_4_4d/n1/b1 | 23.321 ± 0.266023 (W 0.107464) | |
| lm_batch_likelihood_sentence_3_12d/n1/b1 | 87.3276 ± 1.57777 (W 0.545784) | |
| lm_batch_likelihood_sentence_4_4d/n1/b1 | 26.8873 ± 0.0706937 (W 0.162786) | |
| str_matrix_chain_multiplication_100/n1/b1 | 12.7199 ± 0.0448044 (W 0.0338162) | |
| str_nw_mera_closed_120/n1/b1 | 1697.11 ± 0.111872 (W 0.954695) | |
| str_nw_mera_open_26/n1/b1 | 1183.81 ± 1.06703 (W 0.720321) | |
| tensornetwork_permutation_focus_step409_316/n1/b1 | 242.254 ± 3.4272 (W 1.85886) | |
| tensornetwork_permutation_light_415/n1/b1 | 237.858 ± 0.013814 (W 0.328326) | |
| case | opt_size (ms) | notes |
|---|---|
| bin_batched_matmul_b32_m64_n64_k64/n1/b1 | 0.14437 ± 2.82843e-05 (W 0.0013362) | |
| bin_elementwise_mul_2048x2048/n1/b1 | 2.15147 ± 0.0048451 (W 0.0434525) | |
| bin_matmul_256/n1/b1 | 0.313295 ± 0.00631871 (W 0.0102947) | |
| bin_outer_product_4096/n1/b1 | 4.2265 ± 0.0967237 (W 0.0501473) | |
| gm_queen5_5_3.wcsp/n1/b1 | 529.906 ± 0.461702 (W 4.24514) | |
| lm_batch_likelihood_brackets_4_4d/n1/b1 | 10.9638 ± 0.0872754 (W 0.0829693) | |
| lm_batch_likelihood_sentence_3_12d/n1/b1 | 26.4494 ± 0.086634 (W 0.537745) | |
| lm_batch_likelihood_sentence_4_4d/n1/b1 | 15.4098 ± 0.131842 (W 0.195181) | |
| str_matrix_chain_multiplication_100/n1/b1 | 6.85411 ± 0.00978848 (W 0.655965) | |
| str_nw_mera_closed_120/n1/b1 | 509.83 ± 0.782011 (W 5.91522) | |
| str_nw_mera_open_26/n1/b1 | 325.078 ± 0.684881 (W 3.0398) | |
| tensornetwork_permutation_focus_step409_316/n1/b1 | 90.4748 ± 0.661445 (W 1.32435) | |
| tensornetwork_permutation_light_415/n1/b1 | 72.2887 ± 3.39912 (W 3.43573) | ! between-start range over 5% |
Repeated process sets describe observed repeatability under the recorded order schedule; they are not a paired baseline/candidate experiment and do not establish equal performance between engines.