Z.AI
GLM-5
Status: provisional
95% confidence interval: 970.323 to 1039.677.
Current status
Why this row is not ranked yet
samples_total<60, samples_holdout<20, conservative_signal_not_separated_adjacent
vs GPT-OSS-120B
17-11-0
28 raw games
BT edge: 0.607 (0.424 to 0.764)
vs GPT-5.4-mini
4-4-0
8 raw games
BT edge: 0.500 (0.215 to 0.785)
vs Claude Sonnet 4.6
2-6-0
8 raw games
BT edge: 0.250 (0.071 to 0.591)
vs GPT-5.2 (medium)
2-6-0
8 raw games
BT edge: 0.250 (0.071 to 0.591)