Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Training Efficiency on Yuan3.0-1T Pre-training Base (train)
Loading...
92.6
TFLOPS
LAEP
60.9216
69.1458
77.37
85.5942
Jan 20, 2026
TFLOPS
Updated 4d ago
Evaluation Results
Method
Method
Links
TFLOPS
LAEP
Total Params=1010B, La...
2026.01
92.6
DeepSeek-V3 Sequence-Wise Auxiliary Loss
Total Params=1142B, La...
2026.01
84.74
LAEP w/o rearrangement
Total Params=1010B, La...
2026.01
82.25
DeepSeek-V3 Sequence-Wise Auxiliary Loss
Total Params=1515B, La...
2026.01
80.82
Mixtral Auxiliary Load Balancing Loss
Total Params=1515B, La...
2026.01
80.36
Base model
Total Params=1515B, La...
2026.01
62.14
Feedback
Search any
task
Search any
task