llama-stack-mirror/docs/source/distributions/k8s-benchmark/perf_test_config.yaml
Eric Huang c3fa3e6333 base: 35 RPS; safety, 75 RPS
# What does this PR do?


## Test Plan
# What does this PR do?


## Test Plan
2025-09-03 05:58:27 -07:00

19 lines
No EOL
462 B
YAML

version: '2'
image_name: perf-test-demo
apis:
- inference
providers:
inference:
- provider_id: vllm-inference
provider_type: remote::vllm
config:
url: ${env.VLLM_URL:=http://localhost:8001/v1}
max_tokens: ${env.VLLM_MAX_TOKENS:=4096}
api_token: ${env.VLLM_API_TOKEN:=fake}
tls_verify: ${env.VLLM_TLS_VERIFY:=false}
models:
- model_id: ${env.INFERENCE_MODEL}
provider_id: vllm-inference
model_type: llm
server:
port: 8322