Deep20Bench
v1.0Previous

Benchmark edition

Edition 1.0 settings

7 subjects5 rounds per subject50 question limit
Question answersYesNoUnknown
What changed
Source
Menu
Round

Round 2

Round 2

Identified Albert Schweitzer in 38 counted questions.

Back to all Albert Schweitzer rounds
Questions
38
Outcome
Solved
Format errors
Errors found View one example
Duration
16m 0s
Round cost
$0.4745 All 39 turns

Format warning

The model sent replies in the wrong format.

The Guesser sent 10 invalid replies. Each one before the question limit used up a counted turn. The model was only reminded of the required format; it was never told whether its question or guess was correct.

Valid format
74.4%
Format errors
10
Turns used by format errors
10

Transcript

Questions and answers.

Turns are shown in the order they happened. Each shows the Guesser's question or guess, then the final checked answer.

Exact Guesser setupSystem message and BEGIN event

The Guesser received these messages at the start of the round. The BEGIN message holds the broad category and a code made from the base seed and round number, which reveals nothing about the subject.

System message
You are the Guesser being evaluated in a Twenty Questions benchmark.

One subject is hidden. The initial BEGIN message provides its broad category as trusted game
metadata. Identify the subject using as few counted questions as possible. ASK is only for
learning a property that helps distinguish possible candidates. Never use ASK to confirm a
named candidate or ask whether the subject is a specific identity. When you are ready to name
a candidate, use GUESS immediately. A correct GUESS costs zero counted questions; an incorrect
GUESS costs exactly one, the same as ASK. A GUESS must provide both a name and a short
identifying description. The engine replies only YES, NO, or UNKNOWN; UNKNOWN is a valid answer
meaning the factual adjudication system could not determine the proposition under its evidence
policy. Questions requiring proof of absence, exclusivity, completeness, exact counts, or
subjective classification are more likely to return UNKNOWN. Prefer direct, positively
verifiable properties when possible.

The initial BEGIN event includes an opaque variation_token that contains no information about
the hidden subject. Do not interpret, repeat, mention, or derive meaning from it. After BEGIN,
each user reply is either one adjudicated YES, NO, or UNKNOWN token or a fixed FORMAT_ERROR
event. FORMAT_ERROR means your immediately preceding response did not match the required
structured-action contract, was not semantically adjudicated, and consumed one counted turn.
Use the formats in that event and try again. Never infer that the attempted question or
guess was right or wrong.

At most 50 questions or rejected guesses are counted. After that limit you receive
one final guess-only opportunity. Format errors also consume counted turns. The provider wire
response is an object with exactly one `result` action. On the final opportunity ASK is
forbidden. Return only the required structured response. Do not include analysis, hidden
reasoning, or additional fields.
BEGIN event
{"category":"person","event":"BEGIN","variation_token":"BKXG3AZI"}

Valid replies are shown in the required format. Rejected replies are shown separately, without provider or call IDs.

Q1
1 · Guesser asks

Is the person male?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person male?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns Yes
Q2
1 · Guesser asks

Is the person currently alive?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person currently alive?"}}
Oracle evidence1 model-reported source

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q3
1 · Guesser asks

Is the person primarily known as a political leader?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known as a political leader?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q4
1 · Guesser asks

Is the person primarily known for contributions to the field of science (e.g., physics, biology, chemistry, mathematics)?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for contributions to the field of science (e.g., physics, biology, chemistry, mathematics)?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q5
1 · Guesser asks

Is the person primarily known for contributions to the entertainment industry (music, film, television)?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for contributions to the entertainment industry (music, film, television)?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q6
1 · Guesser asks

Is the person primarily known for contributions to literature (novels, poetry, plays)?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for contributions to literature (novels, poetry, plays)?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q7
1 · Guesser asks

Is the person primarily known for contributions to sports (e.g., athlete, coach, or sports figure)?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for contributions to sports (e.g., athlete, coach, or sports figure)?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q8
1 · Guesser asks

Is the person primarily known for contributions to visual arts (painting, sculpture, photography, etc.)?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for contributions to visual arts (painting, sculpture, photography, etc.)?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q9
1 · Guesser asks

Is the person primarily known for their role in a major world religion?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for their role in a major world religion?"}}
2 · Answer check returns Unknown
Q10
1 · Guesser asks

Is the person primarily known for founding or leading a major corporation or business?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for founding or leading a major corporation or business?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q11
1 · Guesser asks

Is the person primarily known for contributions to philosophy?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for contributions to philosophy?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q12
1 · Guesser asks

Is the person primarily known for being a military commander or general?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for being a military commander or general?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q13
1 · Guesser asks

Is the person primarily known for being an explorer or adventurer?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for being an explorer or adventurer?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q14
1 · Guesser asks

Did the person found a major world religion?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Did the person found a major world religion?"}}
2 · Answer check returns Unknown
Q15
1 · Guesser asks

Is the person primarily known as a religious prophet or spiritual teacher?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known as a religious prophet or spiritual teacher?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q16
1 · Guesser asks

Is the person primarily known for committing murders?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for committing murders?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q17
1 · Guesser asks

Is the person primarily known for their contributions to law or jurisprudence?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for their contributions to law or jurisprudence?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q18
1 · Guesser asks

Is the person primarily known for being an engineer or inventor?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for being an engineer or inventor?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q19
1 · Guesser asks

Is the person primarily known as a saint in a major world religion?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known as a saint in a major world religion?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q20
1 · Guesser asks

Is the person known for practicing and advocating Satyagraha (principles of non‑violent resistance)?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person known for practicing and advocating Satyagraha (principles of non‑violent resistance)?"}}
2 · Answer check returns Unknown
Q21
1 · Guesser asks

Is the person primarily known for contributions to economics?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for contributions to economics?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q22
1 · Guesser asks

Did the person receive a Nobel Peace Prize?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Did the person receive a Nobel Peace Prize?"}}
Oracle evidence1 model-reported source

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns Yes
Q23
1 · Guesser reply rejectedTurn counted

Reply in the wrong format.

Output limit exceeded · fixed format reminder sent

Why this reply was rejectedOutput limit exceeded · turn 23

The reply hit the output length limit before it was complete.

Recorded type
output_limit_exceeded
Turn
23
Counted
Yes · question 23
Feedback
FORMAT_ERROR
What the model returnedExact text from the provider

The provider returned no text.

What a valid reply looks likeThe same formats shown to the Guesser
ASK
{"result":{"action":"ASK","description":null,"name":null,"question":"<non-empty yes-or-no property question>"}}
GUESS
{"result":{"action":"GUESS","description":"<non-empty identifying description>","name":"<non-empty candidate name>","question":null}}

A reply must match one of these formats exactly. Text fields in use must not be empty; unused fields must be JSON null.

2 · System responseFORMAT
Q24
1 · Guesser asks

Is the person primarily known for founding a major humanitarian organization?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for founding a major humanitarian organization?"}}
2 · Answer check returns Unknown
Q25
1 · Guesser asks

Did the person serve as Secretary-General of the United Nations?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Did the person serve as Secretary-General of the United Nations?"}}
Oracle evidence1 model-reported source

The Oracle reported these excerpts and URLs. They are not independently checked.

Source 01
Former Secretaries-General: Ban Ki-moon, Kofi Annan, Boutros Boutros-Ghali, Javier Perez de Cuellar, Kurt Waldheim, U Thant, Dag Hjalmar Agne Carl Hammarskjöld, Trygve Halvdan Lie.
Open un.org
2 · Answer check returns No
Q26
1 · Guesser asks

Did the person co‑found the International Committee of the Red Cross (ICRC)?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Did the person co‑found the International Committee of the Red Cross (ICRC)?"}}
Oracle evidence1 model-reported source

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q27
1 · Guesser reply rejectedTurn counted

Reply in the wrong format.

Output limit exceeded · fixed format reminder sent

Why this reply was rejectedOutput limit exceeded · turn 27

The reply hit the output length limit before it was complete.

Recorded type
output_limit_exceeded
Turn
27
Counted
Yes · question 27
Feedback
FORMAT_ERROR
What the model returnedExact text from the provider

The provider returned no text.

What a valid reply looks likeThe same formats shown to the Guesser
ASK
{"result":{"action":"ASK","description":null,"name":null,"question":"<non-empty yes-or-no property question>"}}
GUESS
{"result":{"action":"GUESS","description":"<non-empty identifying description>","name":"<non-empty candidate name>","question":null}}

A reply must match one of these formats exactly. Text fields in use must not be empty; unused fields must be JSON null.

2 · System responseFORMAT
Q28
1 · Guesser reply rejectedTurn counted

Reply in the wrong format.

Output limit exceeded · fixed format reminder sent

Why this reply was rejectedOutput limit exceeded · turn 28

The reply hit the output length limit before it was complete.

Recorded type
output_limit_exceeded
Turn
28
Counted
Yes · question 28
Feedback
FORMAT_ERROR
What the model returnedExact text from the provider

The provider returned no text.

What a valid reply looks likeThe same formats shown to the Guesser
ASK
{"result":{"action":"ASK","description":null,"name":null,"question":"<non-empty yes-or-no property question>"}}
GUESS
{"result":{"action":"GUESS","description":"<non-empty identifying description>","name":"<non-empty candidate name>","question":null}}

A reply must match one of these formats exactly. Text fields in use must not be empty; unused fields must be JSON null.

2 · System responseFORMAT
Q29
1 · Guesser reply rejectedTurn counted

Reply in the wrong format.

Output limit exceeded · fixed format reminder sent

Why this reply was rejectedOutput limit exceeded · turn 29

The reply hit the output length limit before it was complete.

Recorded type
output_limit_exceeded
Turn
29
Counted
Yes · question 29
Feedback
FORMAT_ERROR
What the model returnedExact text from the provider

The provider returned no text.

What a valid reply looks likeThe same formats shown to the Guesser
ASK
{"result":{"action":"ASK","description":null,"name":null,"question":"<non-empty yes-or-no property question>"}}
GUESS
{"result":{"action":"GUESS","description":"<non-empty identifying description>","name":"<non-empty candidate name>","question":null}}

A reply must match one of these formats exactly. Text fields in use must not be empty; unused fields must be JSON null.

2 · System responseFORMAT
Q30
1 · Guesser reply rejectedTurn counted

Reply in the wrong format.

Output limit exceeded · fixed format reminder sent

Why this reply was rejectedOutput limit exceeded · turn 30

The reply hit the output length limit before it was complete.

Recorded type
output_limit_exceeded
Turn
30
Counted
Yes · question 30
Feedback
FORMAT_ERROR
What the model returnedExact text from the provider

The provider returned no text.

What a valid reply looks likeThe same formats shown to the Guesser
ASK
{"result":{"action":"ASK","description":null,"name":null,"question":"<non-empty yes-or-no property question>"}}
GUESS
{"result":{"action":"GUESS","description":"<non-empty identifying description>","name":"<non-empty candidate name>","question":null}}

A reply must match one of these formats exactly. Text fields in use must not be empty; unused fields must be JSON null.

2 · System responseFORMAT
Q31
1 · Guesser asks

Is the person from the United States?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person from the United States?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q32
1 · Guesser reply rejectedTurn counted

Reply in the wrong format.

Output limit exceeded · fixed format reminder sent

Why this reply was rejectedOutput limit exceeded · turn 32

The reply hit the output length limit before it was complete.

Recorded type
output_limit_exceeded
Turn
32
Counted
Yes · question 32
Feedback
FORMAT_ERROR
What the model returnedExact text from the provider

The provider returned no text.

What a valid reply looks likeThe same formats shown to the Guesser
ASK
{"result":{"action":"ASK","description":null,"name":null,"question":"<non-empty yes-or-no property question>"}}
GUESS
{"result":{"action":"GUESS","description":"<non-empty identifying description>","name":"<non-empty candidate name>","question":null}}

A reply must match one of these formats exactly. Text fields in use must not be empty; unused fields must be JSON null.

2 · System responseFORMAT
Q33
1 · Guesser reply rejectedTurn counted

Reply in the wrong format.

Output limit exceeded · fixed format reminder sent

Why this reply was rejectedOutput limit exceeded · turn 33

The reply hit the output length limit before it was complete.

Recorded type
output_limit_exceeded
Turn
33
Counted
Yes · question 33
Feedback
FORMAT_ERROR
What the model returnedExact text from the provider

The provider returned no text.

What a valid reply looks likeThe same formats shown to the Guesser
ASK
{"result":{"action":"ASK","description":null,"name":null,"question":"<non-empty yes-or-no property question>"}}
GUESS
{"result":{"action":"GUESS","description":"<non-empty identifying description>","name":"<non-empty candidate name>","question":null}}

A reply must match one of these formats exactly. Text fields in use must not be empty; unused fields must be JSON null.

2 · System responseFORMAT
Q34
1 · Guesser asks

Is the person primarily known for founding a major humanitarian organization?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Is the person primarily known for founding a major humanitarian organization?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns No
Q35
1 · Guesser reply rejectedTurn counted

Reply in the wrong format.

Output limit exceeded · fixed format reminder sent

Why this reply was rejectedOutput limit exceeded · turn 35

The reply hit the output length limit before it was complete.

Recorded type
output_limit_exceeded
Turn
35
Counted
Yes · question 35
Feedback
FORMAT_ERROR
What the model returnedExact text from the provider

The provider returned no text.

What a valid reply looks likeThe same formats shown to the Guesser
ASK
{"result":{"action":"ASK","description":null,"name":null,"question":"<non-empty yes-or-no property question>"}}
GUESS
{"result":{"action":"GUESS","description":"<non-empty identifying description>","name":"<non-empty candidate name>","question":null}}

A reply must match one of these formats exactly. Text fields in use must not be empty; unused fields must be JSON null.

2 · System responseFORMAT
Q36
1 · Guesser reply rejectedTurn counted

Reply in the wrong format.

Output limit exceeded · fixed format reminder sent

Why this reply was rejectedOutput limit exceeded · turn 36

The reply hit the output length limit before it was complete.

Recorded type
output_limit_exceeded
Turn
36
Counted
Yes · question 36
Feedback
FORMAT_ERROR
What the model returnedExact text from the provider

The provider returned no text.

What a valid reply looks likeThe same formats shown to the Guesser
ASK
{"result":{"action":"ASK","description":null,"name":null,"question":"<non-empty yes-or-no property question>"}}
GUESS
{"result":{"action":"GUESS","description":"<non-empty identifying description>","name":"<non-empty candidate name>","question":null}}

A reply must match one of these formats exactly. Text fields in use must not be empty; unused fields must be JSON null.

2 · System responseFORMAT
Q37
1 · Guesser reply rejectedTurn counted

Reply in the wrong format.

Output limit exceeded · fixed format reminder sent

Why this reply was rejectedOutput limit exceeded · turn 37

The reply hit the output length limit before it was complete.

Recorded type
output_limit_exceeded
Turn
37
Counted
Yes · question 37
Feedback
FORMAT_ERROR
What the model returnedExact text from the provider

The provider returned no text.

What a valid reply looks likeThe same formats shown to the Guesser
ASK
{"result":{"action":"ASK","description":null,"name":null,"question":"<non-empty yes-or-no property question>"}}
GUESS
{"result":{"action":"GUESS","description":"<non-empty identifying description>","name":"<non-empty candidate name>","question":null}}

A reply must match one of these formats exactly. Text fields in use must not be empty; unused fields must be JSON null.

2 · System responseFORMAT
Q38
1 · Guesser asks

Was the person a physician?

Recorded Guesser outputRequired JSON format
{"result":{"action":"ASK","description":null,"name":null,"question":"Was the person a physician?"}}
Oracle evidence2 model-reported sources

The Oracle reported these excerpts and URLs. They are not independently checked.

2 · Answer check returns Yes
G
1 · Guesser guessesNot counted

Albert Schweitzer

German‑French physician, humanitarian missionary, and Nobel Peace Prize laureate

Validator rationale The proposed identity names Albert Schweitzer and its description matches the trusted person.

Recorded Guesser outputRequired JSON format
{"result":{"action":"GUESS","description":"German‑French physician, humanitarian missionary, and Nobel Peace Prize laureate","name":"Albert Schweitzer","question":null}}
2 · Validator returns Yes