Official run

GPT-6.1 SolHigh

In this Twenty Questions LLM benchmark, GPT-6.1 Sol (high) ranks 5 of 18 with a 15.23 question score and 97% success. Lower is better. The run contains 30 scored episodes across 10 subjects with 3 trials per subject. Choose a subject to inspect its attempts.

Rank 5 of 18openaiHigh

Question score15.23questions · lower is better95% CI 12.84–17.62

Success: 97%
Contract compliance: 100%
Guesser cost: $1.13
Guesser time: 40m 2s

Drill down

Subjects.

01Albert Einstein10.00 avg 02Albert Schweitzer19.33 avg 03Garfield9.67 avg 04Achilles8.33 avg 05Genghis Khan10.00 avg 06Bike pump20.67 avg 07Spider web17.67 avg 08Eyebrow14.33 avg 09Moon9.00 avg 10Door handle33.33 avg

Complete benchmark

Run ledger.

All Guesser and benchmark-support activity.

Total tokens: 3,298,775
Wall-clock runtime: 8h 6m
Guesser calls: 486
Completed: Sep 30, 2026, 2:58:08 AM

Reliability

Output contract clean.

All 486 evaluated outputs matched the public structured-action contract.

Run details

Execution: BX-20260929-B-0003-official-M0031-001
Model ID: M-0031
Benchmark: B-0003
Base seed: 0
Git commit: 4226b3254775

Recorded cost

By role.

Cost composition.Share of full-run cost

Guesser39.3%$1.13
Primary Oracle28.8%$0.83
Reviewer15.5%$0.44
Judge16.2%$0.47
Validator0.2%$0.01

Excluded repair overhead$0.543 superseded attempts across 2 affected trials. Not included above.

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-6.1-sol

Asks the questions and submits the scored guess.

Calls: 486
Cost: $1.1275
Reasoning: High
Routing: Exact provider · openai
Resolved model: openai/gpt-6.1-sol
Resolved provider: OpenAI
Prompt contract: stateful-category-guesser-v16-five-answer-category-guide
Configuration: M-0031

Resolved provider details

OpenAI486 calls$1.12752,402.31 s

Fallback calls: 0
Provider unreported: 0

Game support

Answering and checking models.

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: 328
Cost: $0.8272
Reasoning: Medium
Routing: Exact provider · openai
Resolved model: openai/gpt-5.6-luna
Resolved provider: OpenAI
Prompt contract: live-web-oracle-v17-labelled-source-context

Resolved provider details

OpenAI328 calls$0.82722,668.79 s

Fallback calls: 0
Provider unreported: 0

Reviewer

google/gemini-3.5-flash-lite

Checks each Oracle YES or NO independently.

Calls: 315
Cost: $0.4439
Reasoning: Medium
Routing: Exact provider · google-ai-studio
Resolved provider: Google AI Studio

Resolved provider details

Google AI Studio315 calls$0.4439584.37 s

Fallback calls: 0
Provider unreported: 0

Judge

anthropic/claude-opus-5

Decides when the Oracle and Reviewer disagree.

Calls: 21
Cost: $0.4657
Reasoning: Medium
Routing: OpenRouter automatic routing
Resolved provider: Claude Platform on AWS

Resolved provider details

Claude Platform on AWS21 calls$0.4657136.09 s

Fallback calls: 0
Provider unreported: 0

Guess Validator

openai/gpt-5.6-luna

Checks a submitted guess against the trusted subject.

Calls: 35
Cost: $0.0068
Reasoning: Medium
Routing: Exact provider · openai
Resolved model: openai/gpt-5.6-luna
Resolved provider: OpenAI
Prompt contract: strict-guess-validator-v2-generic-kinds
Configuration: gpt-5.6-luna-validator

Resolved provider details

OpenAI35 calls$0.006863.16 s

Fallback calls: 0
Provider unreported: 0