jandres/GenAIExamples
1
0
Fork 0
Code Issues Pull Requests Actions 5 Packages Projects Releases Wiki Activity

fixed visualqna image issues & tgi params issues

Browse Source
This commit is contained in:
Zhenzhong1
2024-10-22 10:26:44 +03:00
parent e21ee76f24
commit f3cbcadfa2
2 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ChatQnA/benchmark/performance/helm_charts/deployment.py
View File

@@ -69,7 +69,7 @@ def generate_yaml(num_nodes, mode="oob", with_rerank="True"):
{ {
"name": "llm-dependency-deploy", "name": "llm-dependency-deploy",
"args": [ "args": [
{"name": "--max-input-length", "values": 1280}, {"name": "--max-input-length", "value": 1280},
{"name": "--max-total-tokens", "value": 2048}, {"name": "--max-total-tokens", "value": 2048},
{"name": "--max-batch-total-tokens", "value": 35536}, {"name": "--max-batch-total-tokens", "value": 35536},
{"name": "--max-batch-prefill-tokens", "value": 4096}, {"name": "--max-batch-prefill-tokens", "value": 4096},

2
VisualQnA/benchmark/performance/helm_charts/values.yaml
View File

@@ -30,7 +30,7 @@ microservices:
- containerPort: 8399 - containerPort: 8399
- name: visualqna-tgi - name: visualqna-tgi
image: opea/llava-tgi:latest image: ghcr.io/huggingface/tgi-gaudi:2.0.5
replicas: 1 replicas: 1
ports: ports:
- containerPort: 8399 - containerPort: 8399