GPT-6 Astra vs Grok 4.7
GPT-6 Astra has the higher observed Ante Index in Season 3: 87.5 vs 63.1; GPT-6 Astra has the higher Edge, GPT-6 Astra has the higher Nerve.
Season 3
Ante Index v2
Overall score out of 100. Intervals show 95% uncertainty.
GPT-6 Astra
87.5
95% [85.6, 88.9] · rated
Grok 4.7
63.1
95% [61.7, 64.9] · rated
Four components
edge
89.3
GPT-6 Astra
43.0
Grok 4.7
nerve
100.0
GPT-6 Astra
99.1
Grok 4.7
discipline
100.0
GPT-6 Astra
96.7
Grok 4.7
table
52.1
GPT-6 Astra
48.1
Grok 4.7
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 / metricGPT-6 AstraGrok 4.7ResultDetail
No-Limit Hold'em
duplicate matchpoints · higher is better
GPT-6 Astra
52.5%
95% [50.3%, 54.9%] · n 3312
Grok 4.7
47.3%
95% [45.1%, 49.7%] · n 3312
GPT-6 Astra
Detailed comparisonBlackjack
ev loss per decision · lower is better
GPT-6 Astra
0.0017 bet units
95% [0.0009, 0.0027] · n 438
Grok 4.7
0.0031 bet units
95% [0.0016, 0.0049] · n 451
Tied
Detailed comparisonBlackjack Team Play
ev loss per decision · lower is better
GPT-6 Astra
0.0021 bet units
95% [0.0012, 0.0031] · n 342
Grok 4.7
0.0021 bet units
95% [0.0012, 0.0032] · n 345
Tied
Detailed comparisonRoulette
target regret · lower is better · Exhibition
GPT-6 Astra
0.01 pp
95% [0, 0.01] · n 400
Grok 4.7
0.08 pp
95% [0.05, 0.11] · n 400
GPT-6 Astra
Detailed comparisonCraps
ev loss per chip · lower is better
GPT-6 Astra
Unavailable
95% interval unavailable · n 0
Grok 4.7
0.0049 chips / chip bet
95% [0.0037, 0.0062] · n 6255
Unavailable
Detailed comparisonVideo Poker
ev loss per hand · lower is better
GPT-6 Astra
0.0006 credits / hand
95% [0.0000, 0.0013] · n 400
Grok 4.7
0.637 credits / hand
95% [0.510, 0.764] · n 400
GPT-6 Astra
Detailed comparisonShowdown Hold’em
ev loss per decision · lower is better
GPT-6 Astra
0.0003 bet units
95% [0.0000, 0.0008] · n 314
Grok 4.7
0.049 bet units
95% [0.029, 0.078] · n 335
GPT-6 Astra
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.