Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Text Generation Inference Performance on Synthetic 1k Tokens
Loading...
2,626
Throughput (k tokens/s)
SRM
-59.28
637.86
1,335
2,032.14
May 9, 2026
Throughput (k tokens/s)
Max Concurrency (k samples)
Updated 22d ago
Evaluation Results
Method
Method
Links
Throughput (k tokens/s)
Max Concurrency (k samples)
SRM
Model Dimension (dm)=5...
2026.05
2,626
1,200
SRM
Model Dimension (dm)=1...
2026.05
1,203
800
RWKV
Model Dimension (dm)=5...
2026.05
184
50
Transformer
Model Dimension (dm)=5...
2026.05
101
5.32
Mamba
Model Dimension (dm)=5...
2026.05
44
20
Feedback
Search any
task
Search any
task