Claude Haiku 4.5 vs GPT-6 Luna
GPT-6 Luna has the higher observed Ante Index in Season 3: 48.6 vs 66.1; GPT-6 Luna has the higher Edge, GPT-6 Luna has the higher Nerve.
Season 3
Ante Index v2
Overall score out of 100. Intervals show 95% uncertainty.
Claude Haiku 4.5
48.6
95% [47.8, 49.3] · rated
GPT-6 Luna
66.1
95% [64.5, 68.6] · rated
Four components
edge
15.1
Claude Haiku 4.5
48.1
GPT-6 Luna
nerve
98.6
Claude Haiku 4.5
99.5
GPT-6 Luna
discipline
98.7
Claude Haiku 4.5
99.8
GPT-6 Luna
table
43.5
Claude Haiku 4.5
47.6
GPT-6 Luna
By game
Each row uses the published headline metric in Season 3. “Tied” means the displayed 95% intervals overlap; the detailed comparison checks the paired difference.
Game / metricClaude Haiku 4.5GPT-6 LunaResultDetail
No-Limit Hold'em
duplicate matchpoints · higher is better
Claude Haiku 4.5
44.4%
95% [42.3%, 46.7%] · n 3456
GPT-6 Luna
47.7%
95% [44.8%, 50.7%] · n 3312
Tied
Detailed comparisonBlackjack
ev loss per decision · lower is better
Claude Haiku 4.5
0.055 bet units
95% [0.044, 0.067] · n 397
GPT-6 Luna
0.0018 bet units
95% [0.0009, 0.0028] · n 438
GPT-6 Luna
Detailed comparisonBlackjack Team Play
ev loss per decision · lower is better
Claude Haiku 4.5
0.045 bet units
95% [0.036, 0.054] · n 329
GPT-6 Luna
0.0020 bet units
95% [0.0010, 0.0032] · n 338
GPT-6 Luna
Detailed comparisonRoulette
target regret · lower is better · Exhibition
Claude Haiku 4.5
1.3 pp
95% [1.04, 1.51] · n 400
GPT-6 Luna
0.78 pp
95% [0.64, 0.93] · n 400
GPT-6 Luna
Detailed comparisonCraps
ev loss per chip · lower is better
Claude Haiku 4.5
0.0078 chips / chip bet
95% [0.0063, 0.0093] · n 5473
GPT-6 Luna
0.0086 chips / chip bet
95% interval unavailable · n 809
Unavailable
Detailed comparisonVideo Poker
ev loss per hand · lower is better
Claude Haiku 4.5
1.200 credits / hand
95% [0.978, 1.452] · n 400
GPT-6 Luna
0.356 credits / hand
95% [0.181, 0.568] · n 400
GPT-6 Luna
Detailed comparisonShowdown Hold’em
ev loss per decision · lower is better
Claude Haiku 4.5
0.597 bet units
95% [0.509, 0.680] · n 454
GPT-6 Luna
0.052 bet units
95% [0.032, 0.075] · n 329
GPT-6 Luna
Detailed comparisonAn interval overlap is a descriptive comparison, not a paired significance test. Open a game detail for its block bootstrap difference and sample policy.