Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
GPU Inference Throughput on A100-SXM4-40GB
Loading...
5,081
Throughput (tok/s)
DF-SSM
-188.68
1,179.41
2,547.5
3,915.59
Apr 28, 2026
Throughput (tok/s)
Speedup
Updated 1mo ago
Evaluation Results
Method
Method
Links
Throughput (tok/s)
Speedup
DF-SSM
Batch=512
2026.04
5,081
-
DF-SSM
Batch=128
2026.04
3,996
-
DF-SSM
Batch=32
2026.04
1,963
4.1
DF-SSM
Batch=8
2026.04
647
5.6
FP16 mamba-ssm
Batch=32
2026.04
482
-
DF-SSM
Batch=1
2026.04
299
21.4
FP16 mamba-ssm
Batch=8
2026.04
116
-
FP16 mamba-ssm
Batch=1
2026.04
14
-
Feedback
Search any
task
Search any
task