Skip to content

run_2026-07-29_nightly

Nightly · v3.2 · 2026-07-29

Claude Sonnet 5.5 · bge-m3 · dense · support-prompt v5 on support-golden-v4 v4.6

CompletedNightlysupport-assistant v3.2v2 · 512 tokens / 64 overlap
Compare

Stratified slice of support-golden-v4 (414 items).

Pass rate

77.1%

−0.9 pts(regression)vs previous nightly319 of 414 samples passed

Faithfulness

0.86

+0.02(improvement)vs previous nightly

Answer correctness

0.82

+0.01(improvement)vs previous nightly

Context precision

0.69

−0.01(regression)vs previous nightly

Context recall

0.76

+0.01(improvement)vs previous nightly

Latency p95

3.6s

+0.1s(regression)vs previous nightly

Failures by category

95 failed samples · largest share: unsupported

Share of failed samples by category, compared with Nightly · v3.2 · 2026-07-28
CategorySharevs reference
Unsupported claims24.2%+0.3 pts
Incorrect factual answers18.9%−0.6 pts
Missing context23.2%+0.3 pts
Retrieval mismatch24.2%+0.3 pts
Instruction-following failures6.3%−0.2 pts
Other3.2%−0.1 pts

Run details

Dataset
support-golden-v4 v4.6 · 414 items
Generator
Claude Sonnet 5.5
Retriever
bge-m3 · dense
Prompt
support-prompt v5
Judge
LLM judge · Claude Opus 5.5 · rubric v3
Started
Jul 29, 02:00 UTC
Duration
43m 12s
Triggered by
Scheduler
Tokens / request
3,870 in · 230 out
  • nightly
  • golden-slice

All metrics

Against Nightly · v3.2 · 2026-07-28

All metrics for Nightly · v3.2 · 2026-07-29
MetricThis runReferenceChange
Pass rate77.1%78.0%−0.9 pts
Faithfulness0.860.84+0.02
Answer correctness0.820.81+0.01
Instruction adherence0.860.860.00
Context precision0.690.70−0.01
Context recall0.760.74+0.01
Recall@50.780.78+0.01
MRR0.650.63+0.02
NDCG@100.730.72+0.01
Latency p501.6s1.6s0.0s
Latency p953.6s3.6s+0.1s
Cost per request$0.0043$0.0044−0.0001

Nightly trend

Nightly runs on the support-golden-v4 slice, Jul 11 to Oct 8. This run is one point in the series.

Pass rate moved from 78.5% to 87.2% across 90 nightly runs. Configuration changes: v3.3 on Aug 4, v3.4 on Sep 8.

Across datasets

support-assistant v3.2 on every dataset in the project. Rows marked “matrix” come from the comparison matrix rather than a dedicated run.

Configuration results per dataset
DatasetItemsPass rateFaith.Ctx recallRecall@5p95Source
support-golden-v4this run1,240
77.1%
0.860.760.783.6sThis run
billing-edge-cases312Not evaluated with this configuration
roaming-and-travel188Not evaluated with this configuration
device-troubleshooting426Not evaluated with this configuration
account-security154Not evaluated with this configuration
locale-formatting96Not evaluated with this configuration
synthetic-hard-negatives540Not evaluated with this configuration

Traced samples

17 samples with full traces (4 failing, 13 passing), stratified by domain. Open a row to inspect retrieval and the judge's reasoning.

Traced samples for Nightly · v3.2 · 2026-07-29

What is the late fee if I pay my bill after the due date?

Billing & payments

Passjudge 0.90

My phone shows 4G but mobile data does not work after the iOS update.

Device troubleshooting

Passjudge 0.87

How do I get a copy of the personal data Harbor holds about me?

Data privacy requests

Passjudge 0.98

Does the Fall 2026 upgrade offer apply if I'm on a family plan?

Promotions

FailMismatchjudge 0.09

Can I return my phone after 20 days if I don't like it?

Returns & warranty

FailIncorrectjudge 0.11

What happens if I go over my fair-use limit on Unlimited at home?

Plans & upgrades

FailMissingjudge 0.07

My eSIM QR code says it has already been used. What should I do?

eSIM & activation

Passjudge 0.87

Is SIM swap protection turned on by default?

Account & security

Passjudge 0.87

Can I return a phone I bought online if I opened the box?

Returns & warranty

Passjudge 0.97

My bill says 41,20 €. Is that forty-one euros or four thousand?

Billing & payments

Passjudge 0.96