Skip to content

run_2026-08-30_nightly

Nightly · v3.3 · 2026-08-30

Claude Sonnet 5.5 · bge-m3 · hybrid (BM25 + dense, RRF) · support-prompt v6 on support-golden-v4 v4.6

CompletedNightlyIn productionsupport-assistant v3.3v2 · 512 tokens / 64 overlap
Compare

Stratified slice of support-golden-v4 (428 items).

Pass rate

81.1%

+0.5 pts(improvement)vs previous nightly347 of 428 samples passed

Faithfulness

0.88

0.00(no change)vs previous nightly

Answer correctness

0.83

−0.01(regression)vs previous nightly

Context precision

0.75

+0.02(improvement)vs previous nightly

Context recall

0.79

−0.01(regression)vs previous nightly

Latency p95

3.9s

+0.2s(regression)vs previous nightly

Failures by category

81 failed samples · largest share: unsupported

Share of failed samples by category, compared with Nightly · v3.3 · 2026-08-29
CategorySharevs reference
Unsupported claims25.9%−0.6 pts
Incorrect factual answers19.8%−0.7 pts
Missing context21.0%+0.5 pts
Retrieval mismatch21.0%+0.5 pts
Instruction-following failures8.6%+0.2 pts
Other3.7%+0.1 pts

Run details

Dataset
support-golden-v4 v4.6 · 428 items
Generator
Claude Sonnet 5.5
Retriever
bge-m3 · hybrid (BM25 + dense, RRF)
Prompt
support-prompt v6
Judge
LLM judge · Claude Opus 5.5 · rubric v3
Started
Aug 30, 02:00 UTC
Duration
46m 4s
Triggered by
Scheduler
Tokens / request
3,980 in · 221 out
  • nightly
  • golden-slice

All metrics

Against Nightly · v3.3 · 2026-08-29

All metrics for Nightly · v3.3 · 2026-08-30
MetricThis runReferenceChange
Pass rate81.1%80.6%+0.5 pts
Faithfulness0.880.870.00
Answer correctness0.830.84−0.01
Instruction adherence0.890.90−0.01
Context precision0.750.73+0.02
Context recall0.790.80−0.01
Recall@50.830.82+0.01
MRR0.700.68+0.02
NDCG@100.760.76−0.01
Latency p501.8s1.9s−0.1s
Latency p953.9s3.7s+0.2s
Cost per request$0.0046$0.0047−0.0001

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 v3.3 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
81.1%
0.880.790.833.9sThis run
billing-edge-cases312
76.0%
0.830.770.813.9sRun · Aug 20
roaming-and-travel188
75.5%
0.840.740.783.8smatrix
device-troubleshooting426
78.9%
0.850.780.824.4smatrix
account-security154
82.5%
0.880.800.843.6smatrix
locale-formatting96
75.0%
0.820.750.794.0smatrix
synthetic-hard-negatives540
70.7%
0.820.680.723.9smatrix

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 · v3.3 · 2026-08-30

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

Plans & upgrades

FailMissingjudge 0.07

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

Returns & warranty

FailIncorrectjudge 0.11

How do I transfer my eSIM to my new iPhone?

eSIM & activation

FailUnsupportedjudge 0.33

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

Promotions

FailMismatchjudge 0.09

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

Billing & payments

Passjudge 0.94

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

Data privacy requests

Passjudge 0.87

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

Returns & warranty

Passjudge 0.97

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

Device troubleshooting

Passjudge 0.87

Is SIM swap protection turned on by default?

Account & security

Passjudge 0.94

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

Billing & payments

Passjudge 0.88