Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Inference Throughput on 64K scenario
Loading...
8,191.3
Prefill Throughput (tok/s)
SparDA
5,224.908
5,995.029
6,765.15
7,535.271
Jun 3, 2026
Prefill Throughput (tok/s)
Throughput (tok/s)
Speedup Ratio
Updated 1mo ago
Evaluation Results
Method
Method
Links
Prefill Throughput (tok/s)
Throughput (tok/s)
Speedup Ratio
SparDA
Backbone=MiniCPM4.1-8B...
2026.06
8,191.3
-
-
Sparse
Backbone=MiniCPM4.1-8B...
2026.06
7,311
-
-
Sparse
Backbone=MiniCPM4.1-8B...
2026.06
7,310.1
-
-
InfiniGen
Backbone=MiniCPM4.1-8B...
2026.06
7,268.6
-
-
Dense
Backbone=NOSA-8B, Hard...
2026.06
6,870.5
-
-
Dense
Backbone=MiniCPM4.1-8B...
2026.06
6,836.1
-
-
SparDA
Backbone=NOSA-8B, Hard...
2026.06
5,817.9
-
-
Sparse
Backbone=NOSA-8B, Hard...
2026.06
5,365
-
-
Sparse
Backbone=NOSA-8B, Hard...
2026.06
5,364.6
-
-
InfiniGen
Backbone=NOSA-8B, Hard...
2026.06
5,339
-
-
gpt-oss-puzzle-88B
Hardware=8x B200, Scen...
2026.02
-
-
1.4
Feedback
Search any
task
Search any
task
