phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-04 12:07:34 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

local

Browse source 
# What does this PR do?


## Test Plan
This commit is contained in:
Eric Huang 2025-09-16 14:13:40 -07:00
parent 3defdf7d3a
commit fe804db21f
2 changed files with 12 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

9
benchmarking/k8s-benchmark/stack-configmap.yaml
View file

@ -5,6 +5,7 @@ data:
image_name: kubernetes-benchmark-demo image_name: kubernetes-benchmark-demo
apis: apis:
- agents - agents
- files
- inference - inference
- files - files
- safety - safety
@ -23,6 +24,14 @@ data:
- provider_id: sentence-transformers - provider_id: sentence-transformers
provider_type: inline::sentence-transformers provider_type: inline::sentence-transformers
config: {} config: {}
files:
- provider_id: meta-reference-files
provider_type: inline::localfs
config:
storage_dir: ${env.FILES_STORAGE_DIR:=~/.llama/distributions/starter/files}
metadata_store:
type: sqlite
db_path: ${env.SQLITE_STORE_DIR:=~/.llama/distributions/starter}/files_metadata.db
vector_io: vector_io:
- provider_id: ${env.ENABLE_CHROMADB:+chromadb} - provider_id: ${env.ENABLE_CHROMADB:+chromadb}
provider_type: remote::chromadb provider_type: remote::chromadb

4
benchmarking/k8s-benchmark/stack-k8s.yaml.template
View file

@ -27,7 +27,7 @@ spec:
spec: spec:
containers: containers:
- name: llama-stack-benchmark - name: llama-stack-benchmark
image: llamastack/distribution-starter:latest image: ehhuang/distribution-starter:latest
imagePullPolicy: Always # since we have specified latest instead of a version imagePullPolicy: Always # since we have specified latest instead of a version
env: env:
- name: ENABLE_CHROMADB - name: ENABLE_CHROMADB
@ -52,6 +52,8 @@ spec:
value: http://vllm-server-safety.default.svc.cluster.local:8001/v1 value: http://vllm-server-safety.default.svc.cluster.local:8001/v1
- name: VLLM_TLS_VERIFY - name: VLLM_TLS_VERIFY
value: "false" value: "false"
- name: LLAMA_STACK_LOGGING
value: "all=WARNING"
command: ["python", "-m", "llama_stack.core.server.server", "/etc/config/stack_run_config.yaml", "--port", "8323"] command: ["python", "-m", "llama_stack.core.server.server", "/etc/config/stack_run_config.yaml", "--port", "8323"]
ports: ports:
- containerPort: 8323 - containerPort: 8323