Official run
Grok 4.6High
In this Twenty Questions AI benchmark, Grok 4.6 (high) ranks 8 of 18 with a question score of 14.29: the average number of questions per round, where lower is better. It solved every round. The run has 35 scored rounds: 7 subjects, 5 rounds each. Choose a subject to see its rounds.
Rank 8 of 18xaiHigh
Round scores are averaged for each subject, then across subjects. A failed round gets the fixed failure score.
The 95% confidence interval shows how certain the average is, based on repeated rounds on these subjects. A wider interval means less consistent results. It does not predict scores on other subjects.
- Solved
- 100%
- Valid format
- 100%
- Guesser cost
- $2.29
- Guesser time
- 1h 49m
Drill down
Subjects.
Complete benchmark
Run totals.
Includes the Guesser and all answer-checking models.
- Total tokens
- 4,936,122
- Total runtime
- 4h 40m
- Guesser calls
- 535
- Completed
Reply format
No format errors.
All 535 checked replies used the required format.
Run details
- Execution
BX-20260814-official-M0015-013- Model ID
M-0015- Benchmark
B-0001- Base seed
0- Git commit
4e197a516062
Recorded cost
By role.
- Guesser27.7%$2.29
- Oracle53.4%$4.41
- Reviewer6.6%$0.54
- Judge12.2%$1.01
- Validator0%$0.00
Roles in this cost:
- Guesser
- The tested model. Asks questions and makes guesses.
- Oracle
- Searches for evidence and proposes an answer.
- Reviewer
- Checks each Oracle answer except Unknown, without seeing it.
- Judge
- Decides when the Oracle and Reviewer disagree.
- Validator
- Checks whether a guess names the hidden subject.
Run configuration
Models.
Models, prompt versions, providers, calls, and cost for the whole run.
x-ai/grok-4.6
The tested model. Asks questions and makes guesses.
- Calls
- 535
- Cost
- $2.2862
- Reasoning
- High
- Routing
- Fixed provider · xai
- Resolved model
- x-ai/grok-4.6
- Resolved provider
- xAI
- Prompt version
stateful-category-guesser-v10-unknown-evidence-guidance- Configuration
M-0015
Provider details
- Fallback calls
- 0
- Provider not reported
- 0
Answer checks
Answering and checking models.
These models answer and check the Guesser's questions and guesses. They are the same for every round and are not being tested.
openai/gpt-5.6-luna
Searches for evidence and proposes an answer.
- Calls
- 499
- Cost
- $4.4071
- Reasoning
- Medium
- Routing
- Fixed provider · openai
- Resolved model
- openai/gpt-5.6-luna
- Resolved provider
- OpenAI
- Prompt version
live-web-oracle-v7-direct-negative-evidence
Provider details
- Fallback calls
- 0
- Provider not reported
- 0
google/gemini-3.5-flash-lite
Checks each Oracle answer except Unknown, without seeing it.
- Calls
- 483
- Cost
- $0.5438
- Reasoning
- Medium
- Routing
- Fixed provider · google-ai-studio
- Resolved provider
- Google AI Studio
Provider details
- Fallback calls
- 0
- Provider not reported
- 0
anthropic/claude-opus-5
Decides when the Oracle and Reviewer disagree.
- Calls
- 81
- Cost
- $1.0093
- Reasoning
- Medium
- Routing
- Chosen automatically by OpenRouter
- Resolved provider
- Amazon Bedrock, Anthropic
Provider details
- Fallback calls
- 0
- Provider not reported
- 0
openai/gpt-5.6-luna
Checks whether a guess names the hidden subject.
- Calls
- 36
- Cost
- $0.0022
- Reasoning
- Medium
- Routing
- Fixed provider · openai
- Resolved model
- openai/gpt-5.6-luna
- Resolved provider
- OpenAI
- Prompt version
strict-guess-validator-v1- Configuration
gpt-5.6-luna-validator
Provider details
- Fallback calls
- 0
- Provider not reported
- 0