Can camelAI show the same answer a different way without changing its values?
camelAI changed its revenue explanation into a chart on request while keeping the underlying answer intact. Both forms showed a 19.1% decline and a $4,383 drop in furniture sales, with the chart using whole-dollar rounding.
1 of 1 test case passed
Every test case in this scenario has a result.
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.
Same answer, different representationPassEvidence
Input 1 · Revenue question: “As of September 2, 2026, how did our revenue do last month, and what drove it?”
Input 2 · Chart request: “Can you show that as a chart instead?”
Response 1 answers Input 1.
Response 2, the final response, answers Input 2.
Response 3 (Chart generation output) is a result the tool showed for Input 2, not a chat message.
AUGUST 2026 / REVENUE SNAPSHOT Revenue fell as order mix shifted down-market Delivered order sales, USD • compared with July 2026 July $10,878 → August $8,795, −$2,082 (−19.1%) Category contribution to the month-over-month change: Furniture −$4,383 · Outdoor & patio −$1,512 · Bedding & rugs +$1,084 · Storage & organization +$960 · Kitchen & dining +$803 · Home décor +$556 · Lighting +$410 Category deltas sum to −$2,082. Cancellations excluded; totals use delivered orders.
Transcribed from Proof 2 · Screenshot
Everything the tool returned, in order (3)
- Revenue answer
August revenue was $8,795.45, down $2,082.30 (19.1%) from July’s $10,877.75. The main driver was lower revenue per delivered order, not fewer orders: • Delivered orders rose from 12 to 22 (+83%). • Average order value fell from $906.48 to $399.79 (−56%). • That shift more than offset the increase in order count. The mix shifted away from high-ticket categories. Furniture revenue fell $4,383 month over month and outdoor & patio fell $1,512. Gains in bedding, kitchen, storage, home décor, and lighting together offset about $3,813 of that decline. Cancellations weren’t the cause of the month-over-month drop: 6 orders were cancelled in each month, and the cancelled order value was lower in August ($5,608 vs. $7,385 in July). Those orders are excluded from revenue.
- Revenue chart
The final response, printed above.
- Chart generation output · not part of the answer
{"ok":true,"stdout":"wrote /outputs/august_2026_revenue_chart.png\n","stderr":"","exitCode":0,"changedFiles":[],"removedFiles":[],"skippedOversize":[],"durationMs":6714}
| ✓ | Expected: Change the presentation without changing the underlying values.Found: Prose totals were July $10,877.75 and August $8,795.45; chart totals were $10,878 and $8,795, rounded. The commerce database returned $10,877.75 and $8,795.45. |
The follow-up asked for a new form, not a new calculation. camelAI returned a chart without prose, preserving the original figures with whole-dollar rounding. A separate query of the commerce database supported those figures.
Sales scope and timing. The revenue explanation identified delivered-order sales rather than total company revenue, noting missing subscription billing data. It also warned that later-dated data meant a period analysis, not a guaranteed snapshot of what was known at the requested date.
Generated chart. camelAI created the revenue chart by writing and running plotting code, rather than retrieving an existing image.
More category detail. The revenue chart added individual contributions for bedding and rugs, storage and organization, kitchen and dining, home décor, and lighting. It stated that the category changes summed to the overall change.
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.
How this scenario is graded
How we decide Pass, Fail and Not gradable. The same rules apply to every tool tested on this scenario.
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.
- 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.
| Level | Name | Scope |
|---|---|---|
| Benchmark | AI Database Agents → | 28 scenarios · 11 tools |
| Capability | Answer Presentation → | |
| Scenario | The user asks to see the same answer a different way → | |
| Tool | camelAI → |
Global scenario definition → · Global capability definition → · camelAI product page →
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.
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.
You run the same kind of test against your own setup and get the same behaviour.
Agree →Yours behaves differently. Tell us what you got, with a screenshot if you have one.
Disagree →Something here is wrong — a reference value, a transcription, a grade.
Report an issue →Tell us what changed and we schedule a rerun of the failing test case. The old result stays as history.
Vendor notice →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.
File fingerprints (SHA-256)
A fingerprint identifies the exact file used for this result.