T3 Code · Performance over time

Process CPU, all threads

All metrics
23 / 120
Process CPU, all threads (lower is better)
102% of one core
18 Nov−1.4% vs the median of up to 7 previous complete nightlies, within noiseFork perf.9 32.6%, −68.7% 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.

CPU time of the whole server process, every thread, over the window.

Lower is better. Noise band 22.2% on Linux x64 (n = 6), so a change counts beyond 27.2%.

Unit
% of one core
Better
lower
Noise band
22.2%, threshold 27.2% (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.cpu.process.busy_pct

Process CPU, all threads (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.20261118.4903102%100 to 1023−1.4%within noise
17 Nov 2026v0.1.11-nightly.20261117.4851105%104 to 1053+1.9%within noise
16 Nov 2026v0.1.10-nightly.20261116.4800103%96.5 to 1043+35.0%regressed beyond noise
15 Nov 2026v0.1.10-nightly.20261115.4749103%101 to 1033+105.4%regressed beyond noise
14 Nov 2026v0.1.10-nightly.20261114.4709105%103 to 1063+108.7%regressed beyond noise
13 Nov 2026v0.1.10-nightly.20261113.4656103%102 to 1053+104.8%regressed beyond noise
12 Nov 2026v0.1.10-nightly.20261112.460677.3%74.9 to 79.73+54.3%regressed beyond noise
11 Nov 2026v0.1.10-nightly.20261111.454454.1%49.6 to 56.83+8.0%regressed beyond noise
10 Nov 2026v0.1.9-nightly.20261110.449154.3%54.1 to 58.93+8.3%regressed beyond noise
9 Nov 2026v0.1.9-nightly.20261109.444349.2%49.1 to 53.13−1.7%within noise
8 Nov 2026v0.1.8-nightly.20261108.438049.2%45.7 to 50.33−1.8%within noise
7 Nov 2026v0.1.8-nightly.20261107.434048.4%48.3 to 48.63−3.5%within noise
6 Nov 2026v0.1.8-nightly.20261106.428849.0%48.8 to 51.73−2.2%within noise
5 Nov 2026v0.1.8-nightly.20261105.423448.4%48.4 to 50.63−3.6%within noise, thin reference
4 Nov 2026v0.1.8-nightly.20261104.419250.2%49.8 to 50.33+0.3%within noise, thin reference
3 Nov 2026v0.1.7-nightly.20261103.413850.1%48.9 to 50.23no reference yet
2 Nov 2026v0.1.7-nightly.20261102.409355.2%54.4 to 55.73no reference yet
1 Nov 2026v0.1.7-nightly.20261101.404857.5%57.4 to 60.63no reference yet
31 Oct 2026v0.1.7-nightly.20261031.400457.2%56.6 to 58.03no reference yet
30 Oct 2026v0.1.6-nightly.20261030.394058.1%57.2 to 59.23no reference yet
29 Oct 2026v0.1.6-nightly.20261029.389246.0%45.2 to 46.23no reference yet
28 Oct 2026v0.1.4-nightly.20261028.384146.3%45.7 to 46.73no reference yet
27 Oct 2026v0.1.4-nightly.20261027.379648.1%47.5 to 48.53no reference yet
26 Oct 2026v0.1.3-nightly.20261026.374049.8%48.2 to 49.93no reference yet
25 Oct 2026v0.1.3-nightly.20261025.368244.0%44.0 to 44.83no reference yet
24 Oct 2026v0.1.3-nightly.20261024.364345.2%44.1 to 45.63no reference yet
23 Oct 2026v0.1.3-nightly.20261023.359246.1%45.4 to 49.93no reference yet
22 Oct 2026v0.1.2-nightly.20261022.354139.1%37.9 to 39.33no reference yet
21 Oct 2026v0.1.2-nightly.20261021.349937.5%36.9 to 37.63no reference yet
20 Oct 2026v0.1.1-nightly.20261020.343837.5%37.4 to 38.23no reference yet
19 Oct 2026v0.1.1-nightly.20261019.337354.7%52.9 to 58.53no noise band yet, no verdict
18 Oct 2026v0.1.1-nightly.20261018.333553.6%52.0 to 54.83no noise band yet, no verdict
17 Oct 2026v0.1.1-nightly.20261017.327252.4%52.3 to 52.53no noise band yet, no verdict
Ladder as a table
AgentsNightly 18 NovFork perf.9
18.50%5.93%
50102%32.6%
10087.7%55.9%
15096.6%71.1%
200107%85.9%
250105%104%
Download this series