Trust & Factuality · sub-racePhoto finish
Honesty & Calibration race
Confidence that matches accuracy, admitting mistakes and resisting sycophancy.
- Leader
- Opus 5.5
- Margin
- 0.4 pts
- 0.4% ahead
- Weeks at #1
- 4
- in a row
- Changes
- 6
- in 2026
Warming up the track…
Oct 19, 2026Opus 5.5 leads
2025Apr ’25Jul ’25Oct ’252026Apr ’26Jul ’26Oct ’26
Score history
Top 6 today over the leader's frontier · click to jump to a week
#1 frontierOpus 5.5Fable 5.1GPT-6.1 SolGPT-6 AstraGemini 4 ArgonSonnet 5.5
Reigns at #1
Longest: GPT-5, 14 weeks (from Aug 2025).
Sibling races
More trust & factuality races
Why this race matters
A model that knows when it is unsure is far safer to build on.
How it is scored
Composite of calibration error and sycophancy resistance.
Unit: percentHigher is betterPhoto finish under 0.5 ptsDominant over 3.0 pts
Data provenance
Calibration & sycophancy evals
Expected calibration error and pushback tests.
Scores here are illustrative placeholders that mimic real benchmark patterns. Swap in live data through the score provider. Methodology