Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
LLM Training on Llama2-70B (64 x H100-8)
Loading...
7.8
Iteration Time (s)
Megatron-LM
7.688
8.444
9.2
9.956
Jul 7, 2025
Iteration Time (s)
MFU
Throughput (tokens/s)
Updated 4d ago
Evaluation Results
Method
Method
Links
Iteration Time (s)
MFU
Throughput (tokens/s)
Megatron-LM
Hardware=64 x H100-8
2025.07
7.8
47.2
-
AXLearn
Hardware=64 x H100-8
2025.07
9.2
40
-
MaxText
Hardware=64 x H100-8
2025.07
9.4
39.1
-
PyTorch FSDP
Hardware=64 x H100-8
2025.07
10.6
34.7
-
Feedback
Search any
task
Search any
task