T3 Code · Performance over time

Agents served in the window

All metrics
2 / 120
Agents served in the window (higher is better)
50
18 Nov0.0% vs the median of up to 7 previous complete nightlies, within noiseFork perf.9 50, 0.0% 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.

Agents the server served in the measured window. Below the level’s agents, the level is saturated: its latencies are shown as context, without verdicts.

Higher is better. Noise band 11.9% on Linux x64 (n = 6), so a change counts beyond 16.9%.

Unit
n
Better
higher
Noise band
11.9%, threshold 16.9% (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
run.json: sessions_served

Agents served in the window (higher 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.20261118.49035050 to 5030.0%within noise
17 Nov 2026v0.1.11-nightly.20261117.48515050 to 5030.0%within noise
16 Nov 2026v0.1.10-nightly.20261116.48005050 to 5030.0%within noise
15 Nov 2026v0.1.10-nightly.20261115.47495050 to 5030.0%within noise
14 Nov 2026v0.1.10-nightly.20261114.47095050 to 5030.0%within noise
13 Nov 2026v0.1.10-nightly.20261113.46565050 to 5030.0%within noise
12 Nov 2026v0.1.10-nightly.20261112.46065050 to 5030.0%within noise
11 Nov 2026v0.1.10-nightly.20261111.45445050 to 5030.0%within noise
10 Nov 2026v0.1.9-nightly.20261110.44915050 to 5030.0%within noise
9 Nov 2026v0.1.9-nightly.20261109.44435050 to 5030.0%within noise
8 Nov 2026v0.1.8-nightly.20261108.43805050 to 5030.0%within noise
7 Nov 2026v0.1.8-nightly.20261107.43405050 to 5030.0%within noise
6 Nov 2026v0.1.8-nightly.20261106.42885050 to 5030.0%within noise
5 Nov 2026v0.1.8-nightly.20261105.42345050 to 5030.0%within noise, thin reference
4 Nov 2026v0.1.8-nightly.20261104.41925050 to 5030.0%within noise, thin reference
3 Nov 2026v0.1.7-nightly.20261103.41385050 to 503no reference yet
2 Nov 2026v0.1.7-nightly.20261102.40935050 to 503no reference yet
1 Nov 2026v0.1.7-nightly.20261101.40485050 to 503no reference yet
31 Oct 2026v0.1.7-nightly.20261031.40045050 to 503no reference yet
30 Oct 2026v0.1.6-nightly.20261030.39405050 to 503no reference yet
29 Oct 2026v0.1.6-nightly.20261029.38925050 to 503no reference yet
28 Oct 2026v0.1.4-nightly.20261028.38415050 to 503no reference yet
27 Oct 2026v0.1.4-nightly.20261027.37965050 to 503no reference yet
26 Oct 2026v0.1.3-nightly.20261026.37405050 to 503no reference yet
25 Oct 2026v0.1.3-nightly.20261025.36825050 to 503no reference yet
24 Oct 2026v0.1.3-nightly.20261024.36435050 to 503no reference yet
23 Oct 2026v0.1.3-nightly.20261023.35925050 to 503no reference yet
22 Oct 2026v0.1.2-nightly.20261022.35415050 to 503no reference yet
21 Oct 2026v0.1.2-nightly.20261021.34995050 to 503no reference yet
20 Oct 2026v0.1.1-nightly.20261020.34385050 to 503no reference yet
19 Oct 2026v0.1.1-nightly.20261019.33735050 to 503no noise band yet, no verdict
18 Oct 2026v0.1.1-nightly.20261018.33355050 to 503no noise band yet, no verdict
17 Oct 2026v0.1.1-nightly.20261017.32725050 to 503no noise band yet, no verdict
Ladder as a table
AgentsNightly 18 NovFork perf.9
111
505050
10058100
150105150
200157200
250153250
Download this series