Official run

GPT-5 NanoMedium

In this Twenty Questions LLM benchmark, GPT-5 Nano (medium) ranks 13 of 18 with a 15.49 question score and 94% success. Lower is better. The run contains 35 scored episodes across 7 subjects with 5 trials per subject. Choose a subject to inspect its attempts.

Rank 13 of 18openaiMedium

Question score15.49questions · lower is better95% CI 13.84–17.13

Success: 94%
Contract compliance: 100%
Guesser cost: $0.48
Guesser time: 3h 14m

Drill down

Subjects.

01Albert Einstein3.80 avg 02Albert Schweitzer44.40 avg 03Stephen King12.00 avg 04Garfield15.60 avg 05Achilles11.60 avg 06Genghis Khan15.60 avg 07Mario5.40 avg

Complete benchmark

Run ledger.

All Guesser and benchmark-support activity.

Total tokens: 5,869,043
Wall-clock runtime: 4h 55m
Guesser calls: 575
Completed: Jul 28, 2026, 9:04:03 PM

Reliability

Output contract clean.

All 575 evaluated outputs matched the public structured-action contract.

Provenance

Execution: BX-20260728-official-M0003-010
Model ID: M-0003
Benchmark: B-0001
Base seed: 0
Git commit: aa131c32c065

Recorded cost

By role.

Cost composition.Share of full-run cost

Guesser10.5%$0.48
Primary Oracle60.7%$2.78
Reviewer12%$0.55
Judge16.4%$0.75
Validator0.5%$0.02

Roles in this cost:

Guesser: Asks the questions and submits the scored guess.
Primary Oracle: Searches for evidence and proposes an answer.
Reviewer: Checks each Oracle YES or NO independently.
Judge: Decides when the Oracle and Reviewer disagree.
Validator: Checks a submitted guess against the trusted subject.

Read the role and answer-checking method →

Run configuration

Models.

Model identity, prompt versions, routing, calls, and recorded cost for the full run.

GuesserModel under test

openai/gpt-5-nano

Asks the questions and submits the scored guess.

Calls: 575
Cost: $0.4839
Reasoning: Medium
Routing: Exact provider · openai
Resolved model: openai/gpt-5-nano
Resolved provider: OpenAI
Prompt contract: stateful-category-guesser-v10-unknown-evidence-guidance
Configuration: M-0003

Resolved provider details

Legacy run. Per-call routing totals were not retained. The configured provider was openai.

Game support

Oracle and adjudication.

These models support the game. They are fixed across the run and are not under test.

Primary Oracle

openai/gpt-5.6-luna

Searches for evidence and proposes an answer.

Calls: 509
Cost: $2.7837
Reasoning: Medium
Routing: Exact provider · openai
Resolved model: openai/gpt-5.6-luna
Resolved provider: OpenAI
Prompt contract: live-web-oracle-v7-direct-negative-evidence

Resolved provider details

Legacy run. Per-call routing totals were not retained. The configured provider was openai.

Reviewer

google/gemini-3.5-flash-lite

Checks each Oracle YES or NO independently.

Calls: 490
Cost: $0.5486
Reasoning: Medium
Routing: Exact provider · google-ai-studio

Resolved provider details

Legacy run. Per-call routing totals were not retained. The configured provider was google-ai-studio.

Judge

anthropic/claude-opus-5

Decides when the Oracle and Reviewer disagree.

Calls: 66
Cost: $0.7511
Reasoning: Medium
Routing: Exact provider · anthropic

Resolved provider details

Legacy run. Per-call routing totals were not retained. The configured provider was anthropic.

Guess Validator

openai/gpt-5.6-luna

Checks a submitted guess against the trusted subject.

Calls: 66
Cost: $0.0210
Reasoning: Medium
Routing: Exact provider · openai
Resolved model: openai/gpt-5.6-luna
Resolved provider: OpenAI
Prompt contract: strict-guess-validator-v1
Configuration: gpt-5.6-luna-validator

Resolved provider details

Legacy run. Per-call routing totals were not retained. The configured provider was openai.