Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Inference Efficiency on Qwen2.5-7B

1,480.2Throughput (tokens/s)

AWQ

-59340.6740.21,139.8Feb 27, 2026Feb 28, 2026Mar 2, 2026Mar 4, 2026Mar 5, 2026Mar 7, 2026Mar 9, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
1,480.20.675-
2026.02
1,093.10.915-
2026.02
1,035.40.966-
2026.02
929.41.0759-
2026.03
27.3-178.3
2026.03
25.6-84.8
2026.03
24-78.5
2026.03
18.9-137.8
2026.03
0.2-50,245.9