Does BlazeSQL generalise a supplied question and SQL pair?
BlazeSQL generalized the supplied question-query pair in this case. It first asked what counted as successful, then after the example query was added it returned 42 for both the repeated question and a reworded one; the same-session reference value was 42.
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.
Verified question/query pair generalisesPassEvidence
Input 1: “As of September 2, 2026, how many successful orders did we have in the last 90 days?”
Input 2: “As of September 2, 2026, how many successful orders did we have in the last 90 days?”
Input 3: “As of September 2, 2026, how many orders were successfully delivered in the last 90 days?”
Response 1 answers Input 1.
Response 2 answers Input 2.
Response 3 is the query the tool generated for Input 2; it is not part of the answer.
Response 4 answers Input 2.
Response 5, the final response, answers Input 3.
Response 6 is the query the tool generated for Input 3; it is not part of the answer.
Response 7, the final response, answers Input 3.
I'll count delivered orders placed in the 90 days ending September 2, 2026. Done—the count is shown above for June 5–September 2, 2026.
Transcribed from Proof 4 · Screenshot
Transcribed from Proof 4 · Screenshot
Every message the tool sent, in order (7)
- Response to input 1
What should count as a successful order for the 90-day period ending September 2, 2026? / Orders placed during the period and delivered by September 2. / Orders placed during the period that were neither cancelled nor refunded as of September 2, including orders still being fulfilled. / Orders delivered during the period, regardless of when they were placed.
- Response to input 2 (message 1 of 3)
I'll count delivered orders placed in the 90-day window ending September 2, 2026, including that date.
- Generated query for input 2 · not part of the answer
SELECT COUNT(*) AS successful_orders FROM orders WHERE status = 'delivered' AND placed_at >= '2026-06-05' AND placed_at < '2026-09-03';
- Response to input 2 (message 3 of 3)
Done—counted delivered orders placed June 5–September 2, 2026. Would you like a monthly breakdown?
- Final response (message 1 of 2)
I'll count delivered orders placed in the 90 days ending September 2, 2026.
- Generated query for input 3 · not part of the answer
SELECT COUNT(*) AS successful_orders FROM orders WHERE status = 'delivered' AND placed_at >= '2026-06-05' AND placed_at < '2026-09-03';
- Final response (message 2 of 2)
Done—the count is shown above for June 5–September 2, 2026.
| ✓ | Expected: Both the repeated question and a rephrasing should match the verified query result; exact wording only would mean lookup-table behaviour.Found: Before training it asked what counted as successful; after training both the repeat and the rephrasing returned 42. |
It passes because BlazeSQL was given one verified question-and-SQL pair through its example-query training, and after that both the exact question and a reworded one returned 42. Before training it asked for a definition instead of answering.
Training changed answer. Training changed the answer: before training it asked for the definition, and after training it answered without asking.
Plain rule wording. Both trained answers stated the rule and date window in plain words.
Example query fields. Example-query training kept the question in the description field and the SQL in the query field.
The before-training screenshot does not show whether a later turn produced a number.
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 | Training → | |
| Scenario | The user supplies example questions and the queries they trust → | |
| Tool | BlazeSQL → |
Global scenario definition → · Global capability definition → · BlazeSQL 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.