Update config.pbtxt

2025-10-20 05:20:39 +00:00 · 2025-10-20 05:20:39 +00:00 · a1841e0300
commit a1841e0300
parent dd8eb6c342
1 changed files with 17 additions and 0 deletions
--- a/config.pbtxt
+++ b/config.pbtxt
@ -130,3 +130,20 @@ instance_group [
  }
 ]

+
+
+#  "model": {
+#    "name": "Llama-3.2-1B-Instruct",
+#    "backend": "TransformerLLM",
+#    "tensorrtllm": {
+#      "workers": 1,
+#      "maxSeqLen": 1,
+#      "kvCacheType": "paged",
+#      "maxInputLen": 1024,
+#      "maxNumTokens": 0
+#    },
+#    "maxBatchSize": 4,
+#    "quantization": "int4",
+#    "modelInstanceGroupKind": "KIND_GPU",
+#    "modelInstanceGroupCount": 1
+#  }