run benchmarks serially

[penelopeysm]

https://aakinshin.net/posts/github-actions-perf-stability/

https://steven-giesel.com/blogPost/77f3667e-cf86-4e62-b09f-b32c0aaf58cd

[github-actions]

Benchmark Report

• this PR's head: 74e9c8bf26ebb029a494496e34618ed1d62e589c
• base branch: 81a245a63bcb9192ffb4ed936e9ca6612137c5e2

Computer Information

Julia Version 1.11.9
Commit 53a02c0720c (2026-02-06 00:27 UTC)
Build Info:
  Official https://julialang.org/ release
Platform Info:
  OS: Linux (x86_64-linux-gnu)
  CPU: 4 × AMD EPYC 9V74 80-Core Processor
  WORD_SIZE: 64
  LLVM: libLLVM-16.0.6 (ORCJIT, znver4)
Threads: 1 default, 0 interactive, 1 GC (on 4 virtual cores)

Benchmark Results

┌───────────────────────┬───────┬─────────────┬────────┬───────────────────────────────┬─────────────────────────────┬───────────────────────────────────┐
│                       │       │             │        │       t(eval) / t(ref)        │      t(grad) / t(eval)      │         t(grad) / t(ref)          │
│                       │       │             │        │ ─────────┬──────────┬──────── │ ───────┬──────────┬──────── │ ──────────┬─────────────┬──────── │
│                 Model │   Dim │  AD Backend │ Linked │     base │  this PR │ speedup │   base │  this PR │ speedup │      base │     this PR │ speedup │
├───────────────────────┼───────┼─────────────┼────────┼──────────┼──────────┼─────────┼────────┼──────────┼─────────┼───────────┼─────────────┼─────────┤
│               Dynamic │    10 │    mooncake │   true │   269.90 │   263.16 │    1.03 │   6.38 │     7.17 │    0.89 │   1721.26 │     1887.11 │    0.91 │
│                   LDA │    12 │ reversediff │   true │  2502.48 │  2388.05 │    1.05 │   2.09 │     2.14 │    0.97 │   5225.75 │     5120.92 │    1.02 │
│   Loop univariate 10k │ 10000 │    mooncake │   true │ 30298.94 │ 28773.63 │    1.05 │   6.77 │     7.10 │    0.95 │ 205194.75 │   204333.84 │    1.00 │
├───────────────────────┼───────┼─────────────┼────────┼──────────┼──────────┼─────────┼────────┼──────────┼─────────┼───────────┼─────────────┼─────────┤
│    Loop univariate 1k │  1000 │    mooncake │   true │  3797.22 │  3798.11 │    1.00 │   5.45 │     5.42 │    1.01 │  20680.67 │    20567.99 │    1.01 │
│      Multivariate 10k │ 10000 │    mooncake │   true │ 40962.92 │ 41711.67 │    0.98 │   6.99 │     6.86 │    1.02 │ 286376.82 │   285979.73 │    1.00 │
│       Multivariate 1k │  1000 │    mooncake │   true │  3948.51 │  4071.77 │    0.97 │   7.89 │     7.58 │    1.04 │  31150.43 │    30845.03 │    1.01 │
├───────────────────────┼───────┼─────────────┼────────┼──────────┼──────────┼─────────┼────────┼──────────┼─────────┼───────────┼─────────────┼─────────┤
│ Simple assume observe │     1 │ forwarddiff │  false │     0.84 │     0.84 │    1.00 │  11.24 │    11.29 │    1.00 │      9.42 │        9.46 │    1.00 │
│           Smorgasbord │   201 │ forwarddiff │  false │   890.18 │   864.41 │    1.03 │  81.69 │ 24473.65 │    0.00 │  72718.49 │ 21155313.82 │    0.00 │
│           Smorgasbord │   201 │      enzyme │   true │  1192.99 │  1203.18 │    0.99 │   4.57 │     4.55 │    1.00 │   5449.62 │     5472.70 │    1.00 │
├───────────────────────┼───────┼─────────────┼────────┼──────────┼──────────┼─────────┼────────┼──────────┼─────────┼───────────┼─────────────┼─────────┤
│           Smorgasbord │   201 │ forwarddiff │   true │  1269.96 │  1639.39 │    0.77 │  76.72 │    57.52 │    1.33 │  97425.73 │    94296.93 │    1.03 │
│           Smorgasbord │   201 │    mooncake │   true │  1233.61 │  1204.12 │    1.02 │   4.49 │     4.57 │    0.98 │   5540.97 │     5505.09 │    1.01 │
│           Smorgasbord │   201 │ reversediff │   true │  1253.56 │  1245.38 │    1.01 │ 119.97 │   120.08 │    1.00 │ 150384.76 │   149545.44 │    1.01 │
├───────────────────────┼───────┼─────────────┼────────┼──────────┼──────────┼─────────┼────────┼──────────┼─────────┼───────────┼─────────────┼─────────┤
│              Submodel │     1 │    mooncake │   true │     0.84 │     0.83 │    1.01 │  30.62 │    33.61 │    0.91 │     25.67 │       27.98 │    0.92 │
└───────────────────────┴───────┴─────────────┴────────┴──────────┴──────────┴─────────┴────────┴──────────┴─────────┴───────────┴─────────────┴─────────┘

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 82.35%. Comparing base (81a245a) to head (74e9c8b).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1368   +/-   ##
=======================================
  Coverage   82.35%   82.35%           
=======================================
  Files          49       49           
  Lines        3502     3502           
=======================================
  Hits         2884     2884           
  Misses        618      618           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[penelopeysm]

That obviously did not help at all with benchmark stability.