Create common performance configs (#22)

Author: dbarbuzzi

common/performance/client.yml

@@ -0,0 +1,12 @@
+data:
+  prompt_tokens: 512
+  prompt_tokens_stdev: 128
+  prompt_tokens_min: 1
+  prompt_tokens_max: 1024
+  output_tokens: 256
+  output_tokens_stdev: 64
+  output_tokens_min: 1
+  output_tokens_max: 1024
+rate-type: sweep
+max-seconds: 400
+warmup-percent: 0.2

common/performance/server.yml

@@ -0,0 +1,5 @@
+uvicorn-log-level: "debug"
+trust-remote-code: true
+enable-chunked-prefill: true
+tensor-parallel-size: 1
+max-model-len: 4096