Update README.md

README.md

@@ -20,5 +20,9 @@ Measured at Wikitext with 4096 context length
 | 5.8438  | 6.9492  |
 
 ## Speed
-Measured with `examples/benchmark_latency.py` script at vLLM repo.
-At batch size = 1, it generates at 13.5 tokens/s with single A100.
+
+Latency and throughput are measured using vLLM (`examples/benchmark_latency.py` and `examples/benchmark_throughput.py` respectively) at single A100-80G.
+
+Latency at batch size 1: 13.5 tokens/s.
+
+Throughput:  0.77 requests/s