T3 Code · Performance over time

SQL statement time per second

All metrics
67 / 120
SQL statement time per second (lower is better)
935ms/s
18 Nov+123.9% vs the median of up to 7 previous complete nightlies, regressed beyond noiseFork perf.9 58.6 ms/s, −85.6% vs its upstream base
AgentsRange

When upstream merges a fix, the lines meet.

Upstream nightlyPerformance forkNoise bandRange of attempts green = better, red = worse, beyond the noise band

As agents grow.

Time spent in SQLite statements per second of the window, reader and writer connections together.

Lower is better. Noise band 42.2% on Linux x64 (n = 6), so a change counts beyond 47.2%.

Unit
ms/s
Better
lower
Noise band
42.2%, threshold 47.2% (n = 6)
Recorded from
17 Oct 2026; before that the series is a gap, never a zero
Platforms
Linux x64
Lane
Every record, inside the agent ladder
Upstream builds
Every upstream build
Source
result.json: (metrics.sql["sqlite|reader"].total_ms + metrics.sql["sqlite|writer"].total_ms) / window seconds

SQL statement time per second (lower is better) at 1 to 250 agents, Linux x64

against
Records, 50 agents
DateTagMedianMin to maxValidvs up to 7 previousVerdict
18 Nov 2026v0.1.11-nightly.20261118.4903935 ms/s920 to 1,0053+123.9%regressed beyond noise
17 Nov 2026v0.1.11-nightly.20261117.4851418 ms/s400 to 4733+0.1%within noise
16 Nov 2026v0.1.10-nightly.20261116.4800375 ms/s293 to 4693−10.4%within noise
15 Nov 2026v0.1.10-nightly.20261115.4749378 ms/s361 to 3933−10.1%within noise
14 Nov 2026v0.1.10-nightly.20261114.4709408 ms/s379 to 4853−6.6%within noise
13 Nov 2026v0.1.10-nightly.20261113.4656458 ms/s456 to 4643+8.9%within noise
12 Nov 2026v0.1.10-nightly.20261112.4606548 ms/s449 to 5603+30.4%regressed beyond noise
11 Nov 2026v0.1.10-nightly.20261111.4544174 ms/s169 to 2013−58.7%improved beyond noise
10 Nov 2026v0.1.9-nightly.20261110.4491264 ms/s198 to 2723−37.1%improved beyond noise
9 Nov 2026v0.1.9-nightly.20261109.4443226 ms/s215 to 2493−46.3%improved beyond noise
8 Nov 2026v0.1.8-nightly.20261108.4380209 ms/s191 to 2413−50.2%improved beyond noise
7 Nov 2026v0.1.8-nightly.20261107.4340213 ms/s211 to 2153−49.4%improved beyond noise
6 Nov 2026v0.1.8-nightly.20261106.4288225 ms/s215 to 2363−46.5%improved beyond noise
5 Nov 2026v0.1.8-nightly.20261105.4234418 ms/s391 to 4333−0.7%within noise, thin reference
4 Nov 2026v0.1.8-nightly.20261104.4192421 ms/s414 to 4373−7.3%within noise, thin reference
3 Nov 2026v0.1.7-nightly.20261103.4138454 ms/s433 to 4653no reference yet
2 Nov 2026v0.1.7-nightly.20261102.4093683 ms/s629 to 7353no reference yet
1 Nov 2026v0.1.7-nightly.20261101.4048728 ms/s713 to 7753no reference yet
31 Oct 2026v0.1.7-nightly.20261031.4004755 ms/s698 to 7863no reference yet
30 Oct 2026v0.1.6-nightly.20261030.3940830 ms/s706 to 8533no reference yet
29 Oct 2026v0.1.6-nightly.20261029.38921,460 ms/s1,459 to 1,4873no reference yet
28 Oct 2026v0.1.4-nightly.20261028.38411,495 ms/s1,422 to 1,5433no reference yet
27 Oct 2026v0.1.4-nightly.20261027.37961,562 ms/s1,546 to 1,5673no reference yet
26 Oct 2026v0.1.3-nightly.20261026.37401,667 ms/s1,549 to 1,7023no reference yet
25 Oct 2026v0.1.3-nightly.20261025.36821,477 ms/s1,338 to 1,4953no reference yet
24 Oct 2026v0.1.3-nightly.20261024.36431,448 ms/s1,361 to 1,4823no reference yet
23 Oct 2026v0.1.3-nightly.20261023.35921,506 ms/s1,488 to 1,6253no reference yet
22 Oct 2026v0.1.2-nightly.20261022.354193.5 ms/s90.2 to 94.03no reference yet
21 Oct 2026v0.1.2-nightly.20261021.349990.2 ms/s84.5 to 93.13no reference yet
20 Oct 2026v0.1.1-nightly.20261020.343893.6 ms/s80.8 to 94.53no reference yet
19 Oct 2026v0.1.1-nightly.20261019.3373667 ms/s637 to 6783no noise band yet, no verdict
18 Oct 2026v0.1.1-nightly.20261018.3335672 ms/s658 to 6993no noise band yet, no verdict
17 Oct 2026v0.1.1-nightly.20261017.3272680 ms/s673 to 6803no noise band yet, no verdict
Ladder as a table
AgentsNightly 18 NovFork perf.9
14.99 ms/s4.61 ms/s
50935 ms/s58.6 ms/s
100245 ms/s143 ms/s
150287 ms/s192 ms/s
200378 ms/s245 ms/s
250395 ms/s306 ms/s
Download this series