tencent/hy4-preview
Tencent
Consistency
Needs 20 scored questions.
How often it gives the same answer when asked again. How it's measured →
Alignment
How often its answer is the one most other models gave. Across 23 questions. How it's measured →
Hedging
moderation 5.6%, blocked by the provider and not counted above
How often it qualifies its answer or declines to give one. How it's measured →
Answers most like
Answers least like
Measured across the questions both models were asked, at least 20 of them. How it's measured →
Where this model stands apart
3 of the 23 scored questions where its answer wasn't the one most models gave.
4 runs
+ 4 responses hide responses
10 runs · consistency 46.7
+ 10 responses hide responses
4 runs
+ 4 responses hide responses
10 runs · consistency 44.4
+ 10 responses hide responses
10 runs · consistency 50
+ 10 responses hide responses
8 runs · consistency 33.3
+ 8 responses hide responses
5 runs
+ 5 responses hide responses
7 runs · consistency 26.7
+ 7 responses hide responses
3 runs
+ 3 responses hide responses
10 runs · consistency not scored
+ 10 responses hide responses
10 runs · consistency 44.4
+ 10 responses hide responses
10 runs · consistency 77.8
+ 10 responses hide responses
10 runs · consistency 27.8
+ 10 responses hide responses
10 runs · consistency 100
+ 10 responses hide responses
8 runs · consistency 66.7
+ 8 responses hide responses
9 runs · consistency 100
+ 9 responses hide responses
8 runs · consistency 40
+ 8 responses hide responses
10 runs · consistency 77.8
+ 10 responses hide responses
10 runs
+ 10 responses hide responses
9 runs