agg-llava.yaml 1.36 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
# SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: Apache-2.0
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
Common:
  model: llava-hf/llava-1.5-7b-hf
  block-size: 64
  max-model-len: 4096

20
21
22
Frontend:
  common-configs: [model]

23
24
Processor:
  router: round-robin
25
  prompt-template: "USER: <image>\n<prompt> ASSISTANT:"
26
27
  common-configs: [model, block-size, max-model-len]

28
VllmDecodeWorker:
29
30
31
  enforce-eager: true
  max-num-batched-tokens: 16384
  enable-prefix-caching: true
32
33
  image-token-id: 32000
  num-patches: 576
34
35
36
37
38
  router: random
  tensor-parallel-size: 1
  ServiceArgs:
    workers: 1
    resources:
39
      gpu: '1'
40
41
  common-configs: [model, block-size, max-model-len]

42
VllmEncodeWorker:
43
44
45
46
47
  tensor-parallel-size: 1
  router: random
  ServiceArgs:
    workers: 1
    resources:
48
      gpu: '1'
49
  common-configs: [model]