agg_router.yaml 1.2 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
# SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: Apache-2.0

apiVersion: nvidia.com/v1alpha1
kind: DynamoGraphDeployment
metadata:
  name: trtllm-agg-router
spec:
  services:
    Frontend:
      dynamoNamespace: trtllm-agg-router
12
      componentType: frontend
13
14
15
16
      replicas: 1
      extraPodSpec:
        mainContainer:
          image: nvcr.io/nvidian/nim-llm-dev/trtllm-runtime:dep-233.17
17
18
19
      envs:
        - name: DYN_ROUTER_MODE
          value: kv
20
21
22
23
24
25
26
27
28
29
30
31
    TRTLLMWorker:
      envFromSecret: hf-token-secret
      dynamoNamespace: trtllm-agg-router
      componentType: worker
      replicas: 2
      resources:
        limits:
          gpu: "1"
      extraPodSpec:
        mainContainer:
          image: nvcr.io/nvidian/nim-llm-dev/trtllm-runtime:dep-233.17
          workingDir: /workspace/components/backends/trtllm
julienmancuso's avatar
julienmancuso committed
32
33
34
          command:
            - /bin/sh
            - -c
35
          args:
julienmancuso's avatar
julienmancuso committed
36
37
            - >-
              python3 -m dynamo.trtllm
38
39
              --model-path Qwen/Qwen3-0.6B
              --served-model-name Qwen/Qwen3-0.6B
julienmancuso's avatar
julienmancuso committed
40
41
              --extra-engine-args engine_configs/agg.yaml
              --publish-events-and-metrics