T3 Code · Performance over time

WAL size, median

All metrics
54 / 120
WAL size, median (lower is better)
41.4MiB
18 Nov+2.7% vs the median of up to 7 previous complete nightlies, within noiseFork perf.9 39.8 MiB, −1.2% 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.

Median size of the SQLite write-ahead log during the window.

Lower is better. Noise band 22.5% on Linux x64 (n = 6), so a change counts beyond 27.5%.

Unit
MiB
Better
lower
Noise band
22.5%, threshold 27.5% (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
proc-1hz.ndjson: wal_bytes median in the window

WAL size, median (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.20261118.490341.4 MiB40.7 to 41.83+2.7%within noise
17 Nov 2026v0.1.11-nightly.20261117.485142.1 MiB41.6 to 42.73+4.9%within noise
16 Nov 2026v0.1.10-nightly.20261116.480041.1 MiB40.8 to 41.33+27.8%regressed beyond noise
15 Nov 2026v0.1.10-nightly.20261115.474940.9 MiB40.5 to 41.23+69.6%regressed beyond noise
14 Nov 2026v0.1.10-nightly.20261114.470940.3 MiB40.1 to 41.23+73.4%regressed beyond noise
13 Nov 2026v0.1.10-nightly.20261113.465640.2 MiB39.8 to 41.23+79.3%regressed beyond noise
12 Nov 2026v0.1.10-nightly.20261112.460687.9 MiB78.1 to 97.83+292.6%regressed beyond noise
11 Nov 2026v0.1.10-nightly.20261111.454473.8 MiB42.7 to 1193+229.5%regressed beyond noise
10 Nov 2026v0.1.9-nightly.20261110.4491105 MiB69.4 to 1213+369.4%regressed beyond noise
9 Nov 2026v0.1.9-nightly.20261109.4443115 MiB94.3 to 1253+414.5%regressed beyond noise
8 Nov 2026v0.1.8-nightly.20261108.4380100.0 MiB78.8 to 1183+346.6%regressed beyond noise
7 Nov 2026v0.1.8-nightly.20261107.4340126 MiB80.9 to 1273+463.1%regressed beyond noise
6 Nov 2026v0.1.8-nightly.20261106.428870.4 MiB36.0 to 89.23+214.4%regressed beyond noise
5 Nov 2026v0.1.8-nightly.20261105.423422.4 MiB12.1 to 26.43−7.2%within noise, thin reference
4 Nov 2026v0.1.8-nightly.20261104.419224.1 MiB15.9 to 26.43+10.7%within noise, thin reference
3 Nov 2026v0.1.7-nightly.20261103.413821.8 MiB19.1 to 28.03no reference yet
2 Nov 2026v0.1.7-nightly.20261102.409320.3 MiB12.2 to 27.53no reference yet
1 Nov 2026v0.1.7-nightly.20261101.404819.4 MiB12.1 to 22.83no reference yet
31 Oct 2026v0.1.7-nightly.20261031.400419.3 MiB19.2 to 24.53no reference yet
30 Oct 2026v0.1.6-nightly.20261030.394026.1 MiB19.6 to 27.13no reference yet
29 Oct 2026v0.1.6-nightly.20261029.389220.2 MiB17.3 to 24.13no reference yet
28 Oct 2026v0.1.4-nightly.20261028.384120.1 MiB16.4 to 24.13no reference yet
27 Oct 2026v0.1.4-nightly.20261027.379625.0 MiB19.1 to 28.93no reference yet
26 Oct 2026v0.1.3-nightly.20261026.374020.9 MiB13.7 to 24.93no reference yet
25 Oct 2026v0.1.3-nightly.20261025.368224.9 MiB13.9 to 25.13no reference yet
24 Oct 2026v0.1.3-nightly.20261024.364324.9 MiB20.3 to 26.63no reference yet
23 Oct 2026v0.1.3-nightly.20261023.359222.2 MiB16.0 to 23.13no reference yet
22 Oct 2026v0.1.2-nightly.20261022.354121.3 MiB21.1 to 29.73no reference yet
21 Oct 2026v0.1.2-nightly.20261021.349930.7 MiB20.5 to 31.23no reference yet
20 Oct 2026v0.1.1-nightly.20261020.343828.7 MiB20.6 to 29.83no reference yet
19 Oct 2026v0.1.1-nightly.20261019.337330.9 MiB30.8 to 323no noise band yet, no verdict
18 Oct 2026v0.1.1-nightly.20261018.333532 MiB32 to 323no noise band yet, no verdict
17 Oct 2026v0.1.1-nightly.20261017.327232 MiB30.8 to 323no noise band yet, no verdict
Ladder as a table
AgentsNightly 18 NovFork perf.9
18.52 MiB7.45 MiB
5041.4 MiB39.8 MiB
10066.7 MiB51.0 MiB
15042.6 MiB42.3 MiB
20041.9 MiB42.1 MiB
25049.2 MiB49.0 MiB
Download this series