From 4ac10ab4f62ec9ebab298b5afd716b928c9d4265 Mon Sep 17 00:00:00 2001 From: Celes Renata Date: Fri, 3 Jul 2026 11:38:24 +0000 Subject: [PATCH] fix: point VLLM_BASE_URL to K8s internal vllm-service vLLM now runs in vllm-service namespace with metrics proxy, accessible at vllm-external.vllm-service.svc.cluster.local:2701 --- infra/helm/stonks-oracle/values-beta.yaml | 2 +- infra/helm/stonks-oracle/values-paper.yaml | 2 +- infra/helm/stonks-oracle/values.yaml | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/infra/helm/stonks-oracle/values-beta.yaml b/infra/helm/stonks-oracle/values-beta.yaml index 1a1678e..abf2cf6 100644 --- a/infra/helm/stonks-oracle/values-beta.yaml +++ b/infra/helm/stonks-oracle/values-beta.yaml @@ -41,7 +41,7 @@ config: BROKER_PROVIDER: "alpaca" OLLAMA_BASE_URL: "http://10.1.1.12:2701" OLLAMA_MODEL: "qwen3.5:4b-fast" - VLLM_BASE_URL: "http://192.168.42.254:2701" + VLLM_BASE_URL: "http://vllm-external.vllm-service.svc.cluster.local:2701" VLLM_MODEL: "numind/NuExtract3" VLLM_TIMEOUT: "120" VLLM_MAX_RETRIES: "2" diff --git a/infra/helm/stonks-oracle/values-paper.yaml b/infra/helm/stonks-oracle/values-paper.yaml index 0040f76..43c7d6b 100644 --- a/infra/helm/stonks-oracle/values-paper.yaml +++ b/infra/helm/stonks-oracle/values-paper.yaml @@ -18,7 +18,7 @@ config: POSTGRES_USER: "stonks_paper" OLLAMA_BASE_URL: "http://10.1.1.12:2701" OLLAMA_MODEL: "qwen3.5:4b-fast" - VLLM_BASE_URL: "http://192.168.42.254:2701" + VLLM_BASE_URL: "http://vllm-external.vllm-service.svc.cluster.local:2701" VLLM_MODEL: "numind/NuExtract3" MARKET_DATA_BASE_URL: "https://api.polygon.io" diff --git a/infra/helm/stonks-oracle/values.yaml b/infra/helm/stonks-oracle/values.yaml index ee1c429..66d959d 100644 --- a/infra/helm/stonks-oracle/values.yaml +++ b/infra/helm/stonks-oracle/values.yaml @@ -192,7 +192,7 @@ config: OLLAMA_RETRY_BASE_DELAY: "1.0" OLLAMA_RETRY_MAX_DELAY: "10.0" OLLAMA_RETRY_BACKOFF_MULTIPLIER: "2.0" - VLLM_BASE_URL: "http://192.168.42.254:2701" + VLLM_BASE_URL: "http://vllm-external.vllm-service.svc.cluster.local:2701" VLLM_MODEL: "numind/NuExtract3" VLLM_TIMEOUT: "120" VLLM_MAX_RETRIES: "2"