Skip to content

Evaluations

Release, comparison and nightly runs for Support Assistant, scored by the rubric v3 judge. Select two to four runs to compare them side by side.

17 release and experiment runs90 nightly runsJudge · Claude Opus 5.5 · rubric v3

Runs, last 30 days

44

1 running · 2 failed

Nightly pass rate

87.2%

Mean of 30 nightly runs

Best full golden-set run

86.9%

support-assistant v3.4 · release candidate

Release gate

Passed

support-assistant v3.4 · release candidate · 7 of 7 checks

107 of 107 runs

Evaluation runs
v3.4 · synthetic-hard-negatives v1.3

Experimentsynthetic-hard-negativesClaude Sonnet 5.5

340 / 540
RunningOct 8
Nightly · v3.4 · 2026-10-08

Nightlysupport-golden-v4Claude Sonnet 5.5

87.2%
Completedp95 4.3s · $0.0041/reqOct 8
v3.4 · Claude Haiku 5.5 comparison

Experimentsupport-golden-v4Claude Haiku 5.5

82.3%
Failedp95 2.3s · $0.0013/reqOct 7
Nightly · v3.4 · 2026-10-07

Nightlysupport-golden-v4Claude Sonnet 5.5

87.3%
Completedp95 3.9s · $0.0041/reqOct 7
v3.4 · Gemini 2.5 Pro comparison

Experimentsupport-golden-v4Gemini 2.5 Pro

83.8%
Completedp95 4.6s · $0.0048/reqOct 6
v3.4 · GPT-5 comparison

Experimentsupport-golden-v4GPT-5

85.3%
Completedp95 5.2s · $0.0061/reqOct 6
support-assistant v3.4 · release candidate

Releasesupport-golden-v4Claude Sonnet 5.5

86.9%
Completedp95 3.8s · $0.0042/reqOct 6
Nightly · v3.4 · 2026-10-06

Nightlysupport-golden-v4Claude Sonnet 5.5

86.6%
Completedp95 3.9s · $0.0042/reqOct 6
Nightly · v3.4 · 2026-10-05

Nightlysupport-golden-v4Claude Sonnet 5.5

86.8%
Completedp95 3.9s · $0.0041/reqOct 5
v3.4 · locale-formatting

Experimentlocale-formattingClaude Sonnet 5.5

80.2%
Completedp95 4.1s · $0.0044/reqOct 4
Nightly · v3.4 · 2026-10-04

Nightlysupport-golden-v4Claude Sonnet 5.5

88.7%
Completedp95 3.9s · $0.0043/reqOct 4
Nightly · v3.4 · 2026-10-03

Nightlysupport-golden-v4Claude Sonnet 5.5

87.9%
Completedp95 3.7s · $0.0042/reqOct 3
v3.4 · billing-edge-cases v2.1

Experimentbilling-edge-casesClaude Sonnet 5.5

81.7%
Completedp95 4.0s · $0.0043/reqOct 2
Nightly · v3.4 · 2026-10-02

Nightlysupport-golden-v4Claude Sonnet 5.5

87.3%
Completedp95 3.6s · $0.0042/reqOct 2
Nightly · v3.4 · 2026-10-01

Nightlysupport-golden-v4Claude Sonnet 5.5

87.6%
Completedp95 3.8s · $0.0041/reqOct 1