AI Demos Research — the structured-intelligence platform. Every verdict on these pages opens to the execution behind it.
Graded 12 November 2026

Does Vectara keep values under the right headers in a simple table?

Vectara correctly answered the row-and-column lookup with 487 for an Extended contract’s out-of-hours callout charge. The answer cited the passage containing the charges table.

1 of 1 test case passed

Every test case in this scenario has a result.

Pass rate100%1 of 1 gradable
Coverage1 of 1test case with a result
1PassDid everything it was expected to do.
0FailNo test case failed.
0Not gradableEvery result here could be graded.
0UntestedEvery test case in this scenario has a result.
The pass rate is a summary. The evidence is the test case below: what we sent, what we checked, what the tool returned, and the proof.

The test case

Each test case is judged on its own: Pass, Fail, Not gradable, or Untested. The scenario result above is based on this test case.

Table value asked by its row and column labelsPassEvidence
What we sent

“What is the out-of-hours callout charge on an Extended cover contract?”

What the tool returned

The out-of-hours callout charge on an Extended cover contract is 487 [1].

Transcribed from Proof 2 · Screenshot

Output file (text)
Text listing: all ten retrieval results with and without reranking
Text93 lines
TC87 — Vectara — R6 Managed RAG v1
Query: What is the out-of-hours callout charge on an Extended cover contract?
Corpus: r6-managed-rag-ss-v1

====================================================================================================
CONFIG 1 — GET /v2/corpora/{corpus_key}/query?query=...&limit=10  (no reranker, tool default)
====================================================================================================

--- rank 1 | score=0.797091 | doc=A01_field_service_policy_and_charges_uk.pdf | page=7 ---
Cover level     Weekday callout     Out-of-hours callout     Offshore day rate
Standard     216     352     1,140
Extended     284     487     1,265
Critical     331     615     1,408

--- rank 2 | score=0.747416 | doc=A05_depot_and_commissioning_note.md | page=None ---
A customer on Standard or Extended cover may borrow a unit where one is free,but has no reservation on it: a Critical cover customer takes precedence and theborrowed unit is recalled with two working days' notice.

--- rank 3 | score=0.744341 | doc=A01_field_service_policy_and_charges_uk.pdf | page=8 ---
Travel
Travel inside a fifty-mile radius of the Bristol workshop is included in the callout charge.

--- rank 4 | score=0.741409 | doc=A01_field_service_policy_and_charges_uk.pdf | page=7 ---
Out-of-hours means any call answered outside 08:00 to 18:00, Monday to Friday, and any call answered on a statutory non-working
day.

--- rank 5 | score=0.735936 | doc=A01_field_service_policy_and_charges_uk.pdf | page=7 ---
A callout is charged once per visit.

--- rank 6 | score=0.723070 | doc=A04_employee_handbook.docx | page=None ---
Anyone rostered to the out-of-hours duty engineer rota is paid a standby allowance for each week on the rota, and an hourly rate for any hour actually worked while on standby.

--- rank 7 | score=0.718964 | doc=A01_field_service_policy_and_charges_uk.pdf | page=7 ---
Callout and attendance charges

--- rank 8 | score=0.708233 | doc=A01_field_service_policy_and_charges_uk.pdf | page=8 ---
Beyond that
radius, travel is charged per mile at the rate in the customer's contract schedule, and any overnight stay is
charged at cost.

--- rank 9 | score=0.707437 | doc=A01_field_service_policy_and_charges_uk.pdf | page=7 ---
How to read this part
The charges below apply to work that falls outside the cover level a customer holds: attendance the
contract does not include, work in the exclusions at section 9, and any attendance a customer requests at
short notice.

--- rank 10 | score=0.700679 | doc=A01_field_service_policy_and_charges_uk.pdf | page=7 ---
Where an engineer attends on two consecutive days, the second day is
charged at the day rate rather than as a second callout.

====================================================================================================
CONFIG 2 — POST /v2/corpora/{corpus_key}/query  (search.limit=10, Vectara default reranker)
====================================================================================================

--- rank 1 | score=0.989672 | doc=A01_field_service_policy_and_charges_uk.pdf | page=7 ---
Cover level     Weekday callout     Out-of-hours callout     Offshore day rate
Standard     216     352     1,140
Extended     284     487     1,265
Critical     331     615     1,408

--- rank 2 | score=0.934634 | doc=A01_field_service_policy_and_charges_uk.pdf | page=7 ---
A callout is charged once per visit.

--- rank 3 | score=0.902438 | doc=A01_field_service_policy_and_charges_uk.pdf | page=7 ---
Where an engineer attends on two consecutive days, the second day is
charged at the day rate rather than as a second callout.

--- rank 4 | score=0.894605 | doc=A01_field_service_policy_and_charges_uk.pdf | page=7 ---
How to read this part
The charges below apply to work that falls outside the cover level a customer holds: attendance the
contract does not include, work in the exclusions at section 9, and any attendance a customer requests at
short notice.

--- rank 5 | score=0.860252 | doc=A05_depot_and_commissioning_note.md | page=None ---
A customer on Standard or Extended cover may borrow a unit where one is free,but has no reservation on it: a Critical cover customer takes precedence and theborrowed unit is recalled with two working days' notice.

--- rank 6 | score=0.859074 | doc=A01_field_service_policy_and_charges_uk.pdf | page=8 ---
Travel
Travel inside a fifty-mile radius of the Bristol workshop is included in the callout charge.

--- rank 7 | score=0.405037 | doc=A04_employee_handbook.docx | page=None ---
Anyone rostered to the out-of-hours duty engineer rota is paid a standby allowance for each week on the rota, and an hourly rate for any hour actually worked while on standby.

--- rank 8 | score=0.388096 | doc=A01_field_service_policy_and_charges_uk.pdf | page=8 ---
Beyond that
radius, travel is charged per mile at the rate in the customer's contract schedule, and any overnight stay is
charged at cost.

--- rank 9 | score=0.375020 | doc=A01_field_service_policy_and_charges_uk.pdf | page=7 ---
Out-of-hours means any call answered outside 08:00 to 18:00, Monday to Friday, and any call answered on a statutory non-working
day.

--- rank 10 | score=0.357792 | doc=A01_field_service_policy_and_charges_uk.pdf | page=7 ---
Callout and attendance charges
Open the text file ↗
Expected vs. Found
✓Expected: The out-of-hours callout charge for Extended cover: 487, taken from the requested row and column.Found: It returned 487; the table’s Extended cover row and out-of-hours column also contain 487.
Supporting proof
Proof 1Screenshot
Notebook screenshot: the question and retrieved charges table without reranking
Open original ↗
Proof 2Screenshot
Notebook screenshot: the generated charge of 487 and its supporting table passage
Open original ↗
Proof 3Screenshot
Notebook screenshot: reranked results and the comparison of top scores
Open original ↗
Why this result

The task required selecting a cell, not merely retrieving its table. Vectara’s generated answer chose the requested row-and-column intersection over all three neighbouring alternatives.

Also observed in this test case

Reranking changes scores. For the top-ranked table passage, reranking raised the score from 0.7971 to 0.9897 without changing its first-place rank. Scores dropped sharply after rank 6.

Answer quality score. The generated answer for the Extended cover out-of-hours charge received a factual-consistency score of 0.738, despite returning the correct value with the correct citation.

Tested by Vivek Tomar · evidence dated 9 September 2026

Configuration and setup

How this tool was set up for the run and what the test needed in place. Each row is a fact from the run's records; a fact the records do not hold is left out, not guessed.

Software that produced the output
Vectara
Build or version
v2
Surface
API
Set up before the run
The field-service charges table had to be ingested with at least two data rows and two labelled columns. The question had to identify the Extended cover row and out-of-hours column, with the charge appearing nowhere else as prose.
Answer generation
enabled with the product's default settings
Tested
9 September 2026 · Vivek Tomar

How this scenario is graded

How we decide Pass, Fail and Not gradable. The same rules apply to every tool tested on this scenario.

How results are decided

Each test case gets one result per tool: Pass, Fail or Not gradable. A test case we haven't run yet shows Untested. There are no partial results.

The rules
  • Pass: the tool did everything the test expected, and nothing it said contradicts the correct answer.
  • Fail: at least one expected behaviour clearly didn't happen; the test case's result says which one and quotes the tool.
  • Not gradable: our evidence couldn't settle the outcome (for example a record we needed is missing). It is never counted as a fail, and the test case's result says what's missing.

Where this sits in the benchmark

This page is one cell of a larger study: one tool, one scenario. Only this benchmark's frame appears here.

LevelNameScope
BenchmarkManaged RAG →15 scenarios · 9 tools
CapabilityData Ingestion →
ScenarioA simple, clearly formatted table →
ToolVectara →

History of this result

What has happened to this result since it was first published. Runs and grades are never overwritten: a retest or a re-grade publishes a new result and keeps the earlier one readable.

from the publication record
13 November 2026First publishedManaged RAG v1

Act on this result

Nothing filed here edits the run or the grade. A challenge opens a review, and a review can produce a new run or a re-grade — which becomes the current result and leaves this one in the history.

This matches what I see

You run the same kind of test against your own setup and get the same behaviour.

Agree →
This does not match

Yours behaves differently. Tell us what you got, with a screenshot if you have one.

Disagree →
Point out an issue

Something here is wrong — a reference value, a transcription, a grade.

Report an issue →
Request a retest

On a newer build, a larger dataset, or your own setup.

Request a retest →
We have fixed this

Tell us what changed and we schedule a rerun of the failing test case. The old result stays as history.

Vendor notice →
Filed against this evidenceNothing yet. Challenges, counter-evidence and fix notices appear here with their outcome, and stay on the page after they are resolved.
Cite this result
aidemos.com/benchmarks/managed-rag/results/vectara/a-simple-clearly-formatted-table · 1 pass · 0 fail · coverage 1/1 · graded 2026-11-12

The same record is available as structured data through the AI Demos MCP server, with the counts, the coverage and every per-test-case reason carried as fields.

Verify the proof files

These files support this result. Open a file to inspect the original evidence.

Proof 1 · Screenshot ↗image/png · 411 KB
Proof 2 · Screenshot ↗image/png · 356 KB
Proof 3 · Screenshot ↗image/png · 200 KB
Output file (text) ↗text/plain; charset=utf-8 · 5 KB
File fingerprints (SHA-256)

A fingerprint identifies the exact file used for this result.

Proof 1 · Screenshotae4500ceb231f3863f9a4ddb99ccf356a1cc63671e696a29ce3d067cde497bc5
Proof 2 · Screenshot3d68b9e81dbf4b4ba87d7166eccfc3753418a94263e43fee018f831d9590aa80
Proof 3 · Screenshotf456feaf26e4194243c0ed542e13ab2f4298c1435d36ef6191d5404cecc2f919
Output file (text)2253d3315d1845a65b8d45dc0356346db7da5414b4dc0a181aeaa2cc0d9b2e11